Skip to main content

Full text of "The Conquest of Khuzistan"

See other formats

The conquest of Khuzistan: 
a historiographical reassessment* 

Oxford University 

In 1889 Ignazio Guidi edited an East Syrian chronicle that covers the late 
Sasanian and very early Islamic period. 1 Four years later Theodor Noldeke 
translated the text into German, dated it to the late seventh century, and 
argued that its provenance was southern, rather than northern Iraq. 2 Noldeke's 
arguments were accepted, and the text came to be called the Khuzistan 
Chronicle, which now seems to be the preferred designation in the secondary 
literature. 3 Little more was said about the text until 1982, 4 when Pierre Nautin 
argued more vigorously for an idea floating around since Noldeke's day, viz. 
that the text consisted of two unequal parts, the second of which was made 
up of what Noldeke called 'notes' (Aufzeichnungen). 5 More specifically, Nautin 
proposed that at least two hands fashioned the work: first a chronicler, who 
he suggested was Elias of Merv (fl. 7th century); 6 and second, at least one (and 
perhaps more) redactor/copyist(s), who added a grab-bag collection of material 
onto the chronicle, which had already lost its beginning; this collection Nautin 
called an 'appendix'. 7 Now whether Elias is to be credited with the first, 

* Versions of this paper were delivered at the Washington meeting of the Middle East Studies 
Association in December 1995, and at the Near and Middle Eastern History Seminar at the 
School of Oriental and African Studies in February 1996. I am indebted to those who listened 
and responded. I am also grateful to Sebastian Brock, Lawrence I. Conrad, and Patricia Crone 
who read and criticized drafts. For several years this article has been described as 'forthcoming 5 
in LI. Conrad, History and historiography in early Islamic times, from which it was reluctantly 
withdrawn. The author regrets any confusion that may result. Abbreviations for periodical and 
other titles are given as follows: AIEO: Annates cie flnstitut d' Etudes Orientates; BE: Byzantinische 
I-orsclnmgen; BGA: Bibliothcca Geographorum Arahicorum; BMGS: Byzantine and Modern Greek 
studies-, CSCO: Corpus Scrip torum Christ ianorum Orientalium; JJS: Journal of Jewish Studies- 
JSS: Journal of Semitic Studies; OCP: Orientalia Christiana Periodica; PdO: Parole de f Orient 
I EQ -Palestine Exploration Quarterly; RHR: Revue de t'histoire des religions; WI: Die Welt 
des Islams. 

' 'Un nuovo testo siriaco sulla storia degli ultimi Sassanidi', Actes du huitieme Congres 
international des Onentalistes (Leiden: E. J. Brill, 1891), Semitics Section, Part B, 1-36 All 
citations here are to Guidi's post-Noldeke edition, Chronicon anonymum, in Chronica Minora 
(Pans: Imprimene nationale, 1903; CSCO 1-2, Scr. syri 1-2), 1, 15-39 (Syriac text)- II 15-32 
(Latin trans.). " ' 

2 Theodor Noldeke 'Die von Guidi herausgegebene syrische Chronik Ubersetzt und 

ll™1893) e \-4S ZW " ^iserlichen Akademie der Wissenschaften, Phil.-Hist. Classe, 

history , BMGS 2 (1976), 23-4; Patricia Crone and Michael Cook, Hagarism: the making of the 
Islamic world (Cambridge: Cambridge University Press, 1977), index of sources, s.v. 'Khnzistan! 
Chronicle ; Ignatius Ortiz de Urbina, Patrologia syriaca (Rome: Pont. Institutum Orientalium 
,n l !ll 0r , l n 1 , 1 ' A 1965 )' 206 -°7; J.-B. Chabot, Litterature syriaque (Paris: Librairie Bloud et Gay, 
1934 103; Anton Baumstark, Geschichte der syrischen Literatur, 2nd ed. (Bonn: A. Marcus und 
b. Webers 1922) 207; Michael Morony, Iraq after the Muslim Conquest (Princeton: Princeton 

V- nn e -!f ty Press '-, 1 , 9 , 84) ' 563< , The Haddad edilion ( noted b y Brock ' 'Syriac historical writing', 
25/302) is unavailable to me, but none of the Ms. variants listed by Brock elsewhere ('Notes on 
some texts in the Mingana Collection', JSS 14 [1969], 221) improves on Guidi's (and Noldeke's) 
readings. v ' 

^-I't^rt? 61 ^ 135 0ne exce P tion: Fie y' s tentative suggestion that either Daniel bar Mariam or 
Mikha o Bet Garme was 'la source ecclesiastique'. See Jean Maurice Fiey, 'Iso'yaw le Grand: vie 
du catnohcos nestonen Iso'yaw III d'Adiabene (580-659)', OCP 36 (1970), 46 n. 3. 

Noldeke, 'Chronik', 2. He also speaks of 'der wenigstens zwei Generationen spSt'er schreibende 
Redactor (ibid, 20 n. 3). 

^On Elias, see Baumstark, Literatur, 208; Chabot, Litterature, 102. 

Mn«Jf ie 4JS ^ autin ' 'L'auteur de la "Chronique Anonyme de Guidi": Elie de Merw', RHR 199 
( Ivo2), 303—14. 

Bulletin of SO AS, 67, I (2004), 14-39. ©School of Oriental and African Studies. Printed in the 
United Kingdom. 



chronicle, section of the work is not at all clear, but Nautin was certainly 
correct to emphasize the contrast between this part and what follows; if any- 
thing pulls the heterogeneous material together here, it is no longer chronology, 
but rather an enthusiasm for geography. 8 

For the date of the composition of the chronicle, Nautin argued for a 
terminus ante quern of 657 or 658, the date of Isho'yab Ill's death; 9 he did not 
date the 'appendix', but much of the evidence cited by Noldeke to date what 
he called a 'letzten Verfassers' would now apply, apparent allusions to the 
conquest of Africa and the failed siege of Constantinople taking us to c. 680. 10 
Noldeke's argument naturally turns on his understanding of these allusions, 
and in fact there are grounds for arguing that Nautili's 'appendix' was com- 
piled even earlier, perhaps very soon after the completion of the chronicle. For 
there are no unambiguous references to events in the 660s and 670s: thus, what 
Noldeke took to be an allusion to the famous siege of Constantinople of the 
late 670s ('Over Constantinople He has not yet given them control') may 
rather allude to obscure events in the 650s. 11 But for our purposes it matters 
little if Nautili's 'appendix' had been compiled by 660, 670, or 680, and I 
shall stick with Noldeke's more conservative dating. 12 The material may have 
been compiled earlier; there is no reason to think that it was compiled later. 

In terms of form and provenance, the 'appendix' is composed of a series 
of discrete accounts, already written in character, 13 and perhaps even more 
clearly than the chronicle, it reflects local knowledge. It is true that similarities 
to material that appears in Monophysite sources suggest that at least some of 
our text's information about Syria came from a Syrian-Byzantine milieu; 14 
but there is precious little of this, and what does come from the West is vague 
in the extreme: there is no doubt that Syria and Egypt were distant places. 
Here it is particularly important to note that unlike much of the later Christian 
tradition that betrays the influence of recognizably Islamic historiographical 
concerns, 15 the 'appendix' — here like the chronicle — shows no reliance on the 
Islamic historical tradition. Entirely absent are features such as Arabic loan 
words (e.g. rasula,fetna), 16 hijn dating, 17 and interests that reflect a specifically 
Islamic Sitz im Leben (e.g. Arabian genealogy). 18 Meanwhile, the names of 

8 It includes, inter alia, an account of one of Elias' miracles, the foundation of several cities 
(see Nautin, 'L'auteur', 307-08), the conquest reports discussed here, Heraclius' death, and some 
Arabian topography. 

9 Nautin, 'L'auteur', 311; Fiey ('Iso'yaw le Grand') puts his death in the year 659. 
10 'Chronik\ 2-3. 

11 As argued by Robert Hoyland, Seeing Islam as others saw it: a study oj the use oj non- 
Muslim sources for early Islamic history (Princeton: Darwin Press, 1996), 185 n. 41. 

12 The text's silence may suggest a date earlier than Noldeke's; it may also reflect the compiler's 
project, since he makes no attempt to be thorough or comprehensive, and is apparently concerned 
to cobble together the stray piece of information that appeals to his interest in geography. 

13 See Noldeke, 'Chronik', 2. 

14 See below, n. 205. 

15 See, for example, Lawrence 1. Conrad, Theophanes and the Arabic historical tradition: 
some indications of intercultural transmission', BF 15 (1990), 1-44. 

16 See the examples adduced in Andrew Palmer, The seventh century in the West-Syrian 
chronicles (Liverpool: Liverpool University Press, 1993), 49 n. 162, 56 n. 173 (rasul); and see also 
the Zuqnin Chronicle, IV, ed. and trans. J.-B.,Chabot under the erroneous title of Chronique tie 
Denys tie Tell Mahre, quatrieme partie (Paris: Emile Bouillon, 1895), ag 967 (fetnd). 

" Such as that in the (West Syrian) Chronicle of 1234; see Jean Maurice Fiey's introduction 
to the French translation of the second volume, Anonymi auctoris chronicon ail annum Christi 
1234 pertinens, II (Louvain: Secretariat du CorpusSCO, 1974; CSCO 354, Scr. syri 154), x; also 
in the East Syrian Opus chronologicum by Elijah of Nisibis (wr. 410/1019), ed. E. W. Brooks 
(Paris: Imprimerie nationale, 1910), 134, where Abu Musa al-Ash'an (whose name is given in full) 
is said to have conquered Bet Huzaye in ah 22. 

18 Such as we have in Theophanes (d. 818); noted by Fred M. Donner, The Early Islamic 
conquests (Princeton: Princeton University Press, 1981), 144. The matter is discussed fully in 
Conrad, Theophanes', 11-16. 



Abu Musa [al-Ash'ari], Khalid [ibn al-Walld], and Sa'd ibn [AbT] Waqqas 
appear in fragmentary form, the Persian general Hormizdan is called 'the 
Mede', and such details as do exist— particularly the names and offices of local 
church notables— are as hard to reconcile with Islamic historiographical 
concerns as they are natural in a local Nestorian Christian milieu. 

It is in the midst of the broadly heterogeneous material in the 'appendix' 
that the reader comes across the subject of this article: a vivid and detailed 
account of the conquest of Bet Huzaye (Ar. Khuzistan/al-Ahwaz). Although 
the Khuzistan Chronicle has been read several times with an eye towards 
discerning a Christian reaction to early Islam in general, 19 it has not yet been 
brought to bear systematically on any of the vexing historical and historio- 
graphical problems that plague students of the conquests. Of course, Noldeke 
did address some of these problems in his translation, but his marginalia are 
spotty and now show their age; 20 in any case, he apparently sought only to 
elucidate the recently available Syriac text. The source has also been put to 
use in a summary of the campaigns of Khalid ibn al-Walld, 21 but there its 
significance lay in its silence about Khalid's presence in Iraq, rather than in 
what it does say about the Muslim presence in Khuzistan. 'As far as the 
conquest is concerned, Islamicists from Wellhausen to Caetani to Donner have 
relied instead on the Arabic sources, and these being generally so intractable, 
and Islamicists generally so conservative, scholarship has hardly moved at 
all. 22 In fact, inasmuch as it has moved, our knowledge has contracted; and it 
is impossible to find fault with Donner's sensible view that' we how must be 
content with 'a sequence of events and with the general understanding that 
the conquest of southern Iraq took place between AD' 635 and 642. To seek 
greater chronological precision is to demand more of the 'sources than they 
can reasonably be expected to provide'. 23 , ' i .■ i !.',y''. 1 ;•:!;•; :\ i ; 

To break the logjam we must leave the Islamic traditiotulnjwhat follows 
I shall do so, putting the long-neglected Syriac text to wOrk by translating and 
commenting on its description of how several cities ih'Klluzistaif ^Fdll ^ to the 
Arabs. 24 My interests are primarily historiographical^'.attd^thbroughly. con- 
ventional at that: I am concerned with the old-fashioned^rVfetillmni&olve^^ 
question of how faithfully our Islamic sources record. conquest history, Of 
course it is impossible to know if the events described ;>by pur andn^mous 
Syriac author actually took place as he describes themV' We- r Gailnot; pretend 
that literary representation, particularly of this variety, is.a i disijri^B|'e^tdd, £ f vVitliess 
to events past, 25 and early sources are not necessarily more Accurate than later 

; .:■: ,-irh f 

19 See, for example, Claude Cahen, 'Note sur l'accueil des chr6ti<ai's"d'Orient k l'lslani\ AHR 
166 (1964), 51-3; Harold Suermann, 'Orientalische Christen •Ufld»tleP -Islam: christliohe Tex te 
aus der Zeit von 632-750', Zeitschrift fiir Missionswissenschafmitid' ReliglonswissenstHaft' 61 
(1983), 130-31; Hoyland, Seeing Islam as others saw it, 182-9. The wofkis curidtlSly abserit from 
M. Benedicte Landron, 'Les relations originelles entre Chretiens de rEst'fnestoriens) et musulmans'. 
PdO 10(1981-82), 191-222. i . ,r]fij| ' 

20 For example, the material on the conquest of Khuzistan- attributed to Sayf ibn 'Umar 
(d. 180/796), and preserved in al-Tabari (wr. 303/915), was not yetavailable to Ndldeke. 

-' See Patricia Crone, art. 'Khalid b. al-Walld' in £7 2 , IV(LeidehJ E. J. Brill, 1978). 961a» <' 
-Julius Wellhausen, 'Prolegomena zur altesten Geschichte'des'Tslams', in his Sklzzen Uhd 
Vorarbeiten, VI (Berlin: Georg Reimer, 1899), 95-6; Leone" Caetani, Amialt deW Islam (Milan: 
Ulnco Hoepli, 1905-26), III, 906-16; IV, 3, 454-74; Donner,' Conquests, 212-17. 

23 Donner, Conquests, 217. • .>nu:o? ■■ib.% w 

24 The 'appendix' also has something to say about matters hi" Syria and Egypt, which I have 
translated in a brief appendix of my own; it follows below. ; -iy-V • 

25 The point hardly needs demonstration, but cf. John'WahsbroUgh, The sectarian milieu: 
content and composition of Islamic salvation history (Oxfordr Oxford tlnivertlty PresS; 1978), 119: 
' ought to be clear that there can be no question of a r neutral or "objective" source"; Each 
witness, regardless of its confessional alignment, exhibits a* similar, if not ttltc ;ether identical 
concern to understand the theodicy'. 



ones. 26 But if we shall never know exactly what happened in Khuzistan in the 
640s and 650s, our Syriac source preserves a very early understanding of what 
happened, and in so doing it provides an invaluable control for the later 
Islamic tradition. Early, naive, and historiographically independent of Islamic 
sources, it allows us to identify and occasionally disentangle strands of tradition 
that are manifestly late and polemically conditioned from other, older, strands 
that preserve authentically early views of conquest history. 

The Syriac account 

The relevant account may be translated as follows. 27 I have broken the text 
into paragraphs for the sake of clarity. 

At the time of which we have been speaking (beh den b-hana zabnci d-men 
I' el emaman), when the Arabs (tayyaye) conquered all the lands of the 
Persians and Byzantines, 28 they also entered and conquered all the fortified 
towns, that is, Bet Lapat (Ar. Jimdaysabur), 29 Karka d-Ledan, 30 and 
Shushan, the citadel. 31 There remained only Shush (Ar. al-Sus) and 
Shushtra (Ar. Tustar), which were very strong, while of all the Persians 
none remained to resist the Arabs except king Yazdgard 32 and one of his 
commanders (had men rabbay haylawateh), whose name was HormTzdan 
the Mede, 33 who gathered troops and held Shush and Shushtra. This 
Shushtra is very extensive and strong, because of the mighty rivers and 
canals that surround it on every side like moats. One of these was called 
ArdashTragan, after Ardashlr who dug it; another, which crossed it, was 
called ShamTram, after the queen; and another, Darayagan, after Darius. 
The largest of all of them was a mighty torrent, which flowed down from 
the northern mountains. 34 

26 It is regrettable that this point is usually made apologetically, in defence of late evidence; 
see K. Lawson Younger, Ancient conquest accounts: a study in ancient Near Eastern and biblical 
history writing (Sheffield: JSOT Press, 1990), 249-53. 

27 The passage begins on 35:20/29:30 and ends at 37:14/31:2. 

28 As Noldeke remarked ('Chronik', 41 n. 4) this passage seems to allude to an earlier one, 
which begins on 30:23/26:13: 'Then God brought the sons of Ismail against them, [innumerable] 
like sand on the sea shore. Muhammad was their leader (mdabbrdna). Neither walls, gates, armor, 
or shields withstood them, and they took control over all of the land of the Persians. Yazdgard 
sent countless armies against them, but the Arabs (tayyaye) defeated them all; they even killed 
Rustam. Yazdgard shut himself up inside the wall's of Mahoze (i.e. Seleucia-Ctesiphon), but 
eventually escaped by fleeing. He came to the lands of the Huzaye and of the Maronaye. There 
he ended his life. The Arabs took control of Mahoze and all of its lands. They also came to the 
Byzantine lands, and they plundered and ravaged all of the lands of Syria. Heraclius, the king of 
the Byzantines, sent armies against them, but the Arabs killed more than 100,000 of them'. 

29 On Bet Lapat, see Jean Maurice Fiey, 'L'Elam, la premiere des metropoles ecclesiastiques 
syriennes orientales', Melto 5 (1969), 227-67; reprinted in idem, Communautes syriaques en Iran 
et Irak des origines a 1552 (London: Vqriorum, 1979), Chapter III. 

30 On Karka d-Ledan, see Fiey, 'L'Elam, la premiere des metropoles ecclesiastiques syriennes 
orientales (suite)', PdO 1 (1970), 123-30; reprinted in his Communautes, Chapter 1 1 lb. 

31 As Noldeke comments ('Chronik', 42 n. 2), the phrase is biblical, but the author clearly 
does not have in mind Shush (Susa, al-SQs), which presently follows. 

32 i.e. Yazdagird III (r. 632-51). 

33 On the name, see Ferdinand Justi, Iranisches Namenbuch (Marburg: N. G. Elwert'sche 
Verlagsbuchhandlung, 1895), 10. The Arabic form preferred is generally al-Hurmuzan, with the 
important exception of Ibn A'tham (wr. 204/819-20), whose reading (H-r-m-z-d-a-n) comes closest 
to the Syriac. On the date and transmission history of Ibn A'tham's history, several recensions of 
which have survived — at least in part — to modern times, see Lawrence 1. Conrad, Ibn A'tham and 
his history (Winona Lake, IN: American Oriental Society, forthcoming). 

34 For a convenient discussion of the region's geography, see W. Barthold, An historical 
geography of Iran, trans. Svat Soucek, ed. C. E. Bosworth (Princeton: Princeton University Press, 
1984), Chapter 11. 



Then {hayden) an Arab commander known as 35 Abu Musa attacked 
HormTzdan the Mede. He (Abu Musa) had built al-Basra as a settlement 
(l-mawffibhon) for the Arabs, where the Tigris flows into the great ocean, 
between the cultivated land and the desert, just as Sa'd bar Waqqas had 
built the city of 'Aqula as another settlement for the Arabs, which was 
named Kufa, after the bend of the Euphrates. But when Abu Musa went 
to attack Hormizdan, this HormTzdan devised stratagems in order to pre- 
vent them (the Arabs) from engaging him, until he gathered an army. He 
wrote to Abu Musa that he (Abu Musa) should stop taking captives and 
making war, and that he (HormTzdan) would send him whatever tribute 
(madatta) they imposed on him. Thus it remained for two years. 

Trusting his walls, HormTzdan then broke the truce (shoyna) between them, 
and killed the men who had been ambassadors between them, one of whom 
was George, the bishop of Ulay. 36 He [also] imprisoned Abraham, the 
metropolitan of Furat. 37 He [then] sent many armies against the Arabs, 
but they defeated them all. The Arabs rushed [forward], lay siege to ShUsh,' 
took it after a few days, and killed all of the nobles (pnslie) in it. They 
seized the house that is called the 'House of Mar Daniel', and took the 
treasure there enclosed, which had been kept there on the kings' orders 
since the days of Darius and Cyrus. They also broke open and made off 
with a silver coffin, in which a mummified corpse was laid; many said it 
was Daniel's, but others [claimed] that it was Darius. 

They also lay siege to Shushtra, and fought for two years to take it. Then 
a man from Qatar 38 who was living there befriended a man who had a 
house on the walls, and the two of them conspired together. They went 
out to the Arabs and told them: 'If you give us a third of the spoil of the 
city, we will let you into it'. They came to an agreement, dug tunnels under 
the walls, and let in the Arabs, who [thus] captured Shushtra. They shed 
blood there as if it were water. They killed the exegete of the city and the 
bishop of HormizdardashTr (Ar. Suq al-Ahwaz), 39 along with the students, 40 
priests, and_ deacons, whose blood they shed in the holy sanctuary. They 
took Hormizdan alive. 

The passage translated appears to be a discrete unit. With a sure command 
of detail, and paced by a series of adverbs and adverbial phrases that link the 
episodes temporally and logically, the account generates a sense of movement 
that is almost entirely lacking in other parts of the 'appendix'. Elsewhere 
information is imparted: here a coherent story is told. Since our .compiler 
generally shows little if any historical method, 41 we can assume that the ac'dount 
came to him in this form; he copied it, just as he copied the chronicle .before 
it. Its appeal presumably lay in the quality of its narrative, which vividly 



Meikne, usually merely 'nicknamed', but here it precisely expresses the Afabicrk'tihW 

' Apparently located south of al-HTra; see Morony, Iraq, 152; Donner, Cduqutitsp329m66. 

I hat is, Furat d-Maysan, which was apparently located opposite the medieval'' site of 

al-Basra; see Morony, Iraq, 159. ' ', „-?, rjf.^.rl 'y. m n ,.] 

■ Noldeke ('Chronik', 25, n. 2) points out that this was understood broadly;: 'Qalattrnfasst 

aber bei dtesen Syrern alle Lander der nordostlichen Arabiens, wo damals Viele^nes'toriahische 

ed and trans. Arthur Voobus (Stockholm: Papers of the Estonian Theolbgical:Sodie#'iii Exile! 

1962) esp. 79; J. B Segal, Edessa, The Blessed City' (Oxford: Clarendon Presg/ir970)/i49^5l 

It is particularly worth noting that no effort has been made to'relatdr^alirn'nial-ch to 
Syria, as portrayed in the 'appendix', to the chronicler's earlier allusion td ttt-tf armflfc foil which 
see below). / t ( -,,.,,, \ ./ r . 



describes the terrible fate of a Nestorian heartland; it may also have appealed 
to the copyist's (or copyists') interest in geography and topography. Whatever 
its ultimate provenance, it is more detailed than anything available to Elias 
of Merv, 42 or, for that matter, anything else to be written in either West or 
East Syriac. 

What the Syriac account cannot tell us 

In what follows I shall argue that the 'appendix' to the Khuzistan Chronicle 
can provide enough corroboration for accounts in the Islamic tradition that 
we must posit the continuous transmission of historical material within the 
latter. In this case, some early material clearly did survive the hazardous 
passage from witness to tradent to historian, a passage of approximately 
150-200 years. The degree to which those who initially transmitted and com- 
piled the material were concerned with what we would consider historio- 
graphical issues — particularly problems of sequence and time — is considerably 
harder to discern, and although we shall meet these problems throughout, it 
is best if we address two at the start. 

First, since our source begins with the entrance of Abu Musa al-Ash'an, 
it sheds no light on the events that the Islamic tradition describes as having 
taken place before his appearance in Khuzistan: of cities that are said to have 
entered into treaties, which they would soon break, and of 'Utba ibn Ghazwan 
and al-Mughlra ibn Shu'ba, the two commanders who are said to have preceded 
Abu Musa on the front, we hear nothing. 43 What our source does say, however, 
is that all but four of the 'fortified towns' had been taken before Abu Musa 
arrived on the scene; and thus there is probably something to the Islamic 
accounts that attribute some role to 'Utba and al-Mughlra. 44 Of course, 
whether Abu Musa's victories can be considered the last phase of a continuous 
series of campaigns that began with 'Utba is altogether a different question, 
and one that the source does not answer: the world of Medinan state building 
and caliphal politics is unknown to our Syriac source. Our Syriac compiler 
was apparently concerned only to record the outlines of the Sasanian defeat, 
rather than a detailed history of the Muslim victory; and even assuming that 
he had heard of such earlier battles as there were, we can hardly expect him 
to have connected them to those led by Abu Musa. He records what the 
Islamic tradition generally considers the final phase of the conquest of 
Khuzistan, probably for the simple reason that Abu Musa's campaigns were 
indeed decisive. 

Although Syriac accounts can occasionally provide invaluable help in 
solving dating problems, 45 this one cannot; here we arrive at the second 
principal limitation of our source. An assortment of topics, 46 the 'appendix' 
can only yield a relative dating, and one that happens to be particularly weak 
to boot. The beginning of the passage suggests that the start of the conquest 

42 Be that in the chronicle part of the work, following Nautin, or in the Christian Arabic 
Chronicle of Seer l, following L. Sako, 'Les sources de la chronique deSeerf, NO 14 (1987), 159. 
On the disputed authorship of this work, see Jean Maurice Fiey, ishd'dnah et La chronique 
Je Seen', PdO 6-7 (1975-76), 447-59; and Naulin's riposte in 'L'auteur', 313-14. 

43 For summaries of these events, see the works cited above, n. 22. 

44 Here it is tempting to infer from the presence of the bishop oi' Hormlzdardashlr in Tustar 
that his city had already fallen. 

45 Particularly for events in Syria and Palestine, where the Christian testimony is most dense; 
the earliest example is Thcodor Noldeke, 'Zur Geschichle der Aruber im 1. Jahrhundert d.H. aus 
syrischen Quellen', ZDMG 29 (1875), 76-98. 

46 In Nautin's words ('L'auteur', 304), 'un appendice fait de morceaux decousus'. 

umx~vj-v r*rr- 

.: ' .i,.u»JillM 

m-uv-j^iufcii-u j-<-jnJ-.*."-ii-?tf 



of Khuzistan was roughly contemporaneous with, or perhaps even followed, 
that of Iraq and Syria: 'At the time of which we have been speaking, when 
the Arabs conquered all the lands of the Persians and Byzantines, they also 
entered and conquered all the fortified towns...'. But after recording Abu 
Musa's campaigns, it then turns to Khalid ibn al-Walld's conquest of Syria, 
which it says followed those of Abu Musa: 'Afterwards (bdtarken) a man from 
the Arabs named Kaled came and went to the West, and took the lands and 
towns as far as 'Arab'. 47 Now the problem can be solved by preferring the 
second of these two passages, which has the virtue of more clearly asserting a 
sequence of events; and since the remarks that follow seem to allude to the 
battle of al-Yarmiik, 48 we can actually generate a terminus ante quern of late 
August of 636/Rajab of ah 15 for the end of Abu Miisa's campaigns. 49 That 
this dating is at severe variance with the consensus of the Islamic sources might 
cause some concern, 50 particularly because it would force a redating of the 
founding of al-Basra; but it is far from fatal, the Islamic tradition containing 
some aberrant dating schemes of its own. A report in the Kitab al-kharaj of 
Abu Yusuf (d. 182/798), for example, can be handled in such a way so as to 
produce the dating of e. ah 15 or 16 for the fall of Tustar. 51 

But there are too many problems to overcome. For one thing the sequence 
of conquests would run afoul of another, earlier, non-Islamic source. 52 For 
another, it is not at all clear that the author of the passages translated above 
can also be credited with the passage translated below; and since the final 
redactor/editor manifests so little interest in chronology, we cannot use the 
latter to date material in the former without establishing single authorship. 
Moreover, even if we could establish a single author, his acquaintance with 
events in Syria pales in comparison with his knowledge of his (apparently) 
native Bet Huzaye; and it would be nothing if not reckless to use his vague 
and secondhand material concerning the West to date his detailed account of 
local events. Finally, it may be that the crucial adverb (batarken)— the hinge 
upon which the proposed dating would swing— has little temporal significance, 
and instead marks nothing more than a narrative transition. 53 


47 For the whole passage, see the Appendix below. 

48 Cf. Noldeke, 'Geschichte', 79; Palmer, Seventh century, 3. 

49 According to the conventional interpretations of M. j! de Goeje, Memoires stir la conquete 
de la Syrie, 2nd ed. (Leiden: E. J. Brill, 1900), 108-24; Donner, Conquests, 128-44; Palmer, 
Seventh century, 4. :•::»,....• .n \ 

50 The earliest date for operations in Khuzistan seems to be the consensus report (qalu, !they 
said') that begins al-Baladhuri's section on al-Ahwaz; but here it is al-Mughlra ibri Shu'ba who 
raids Siiq al-Ahwaz in late 15 or early 16/636 or 637; see al-Baladhurl (d\*fflm92Jl?Pitffih 
al-huldan, ed. ML J. de Goeje (Leiden: E. J. Brill, 1866), 376. Khalifa ibn KMyyat (d, .240/854)', 
Ta'iikh, ed. Suhayl Zakkar (Damascus: Wizarat al-thaqafa wa-1-siyaha wa-1-irstiad al-qawml, 
1967), I, 105, puts this raid in ah 16. The latest date is in the severely telescoped' account in 
al-Ya'qiibT (d. 284/897), Historiae, ed. M. T. Houtsma (Leiden: E. J. Brill, 1883), II, 180,- where 
Abu Musa's conquest of al-Ahwaz and Istakhr is put in ah 23. • u ,'•.;•;/ .'?rr- ,-,, 

51 The report states that Abu Musti conquered Tustar, Isfahan, Mihrajanqadhaq, and 
Nihawand (?) while Sa'd ibn AbT Waqqas was laying siege to al-Mada'in;'see ,, Abu' Yustlf, Kitab 
al-kharaj (Cairo: al-Matba'a al-salafiya, ah 1352), 60. The date for the final capitulation of 
al-Mada'in is usually given as 16/637; see al-Tabarl, Ta'rtkh al-rusul wa-l-muluk, ed.-M. J. de Goeje 
et al, (Leiden: E. J. Brill, 1879-1901), I, 2431-32. But its siege pay have been Very; protracted; 
al-Dmawarl (d. 282/891), Al-Akhhar al-tiwal, ed. Vladimir Guirgass andTgnatius:Kfatchkovsky 
(Leiden: E. J. Brill, 1888-1912), 133, puts it at 28 months. See also al^BaladhUfi^ft/ty 262-64; 
Yaqut (d. 626/1229), Mu'jatn al-buhlan, ed. Ferdinand Wustenfeld (Leipzig: VFt> A^firdckhaus, 
1866-73), I, 768, which draws on al-BaladhurT, as well as on a chronology that tiate^the^ohquest 
to ah 15. ,- ■■ ■• . |,,i,..„-' •■ ; . ,-M 

-See ps.-Sebeos (wr. c. 660-70), Hist aire ti'Heraclius, trans. Frederid'Ma'de^CParis: fernest 
Leroux, 1904), 97-101. , .}.• . ;. :v . : , ,. . i;: v^; :' 

53 For parallels in the Arab-Islamic tradition, see Albrecht NothyTA* \eatly. ArabiCihlstmcal 
tradition: a source-critical study, 2nd ed. in collaboration with Lawrenceli'GShradjItrahB.i.M'tchael 
Bonner (Princeton: Darwin Press, 1994; hereafter cited as Noth/Corira'dV 173*77. • '"V: n\ '" 



In sum, nothing in the 'appendix' can yield a precise date for the conquest 
of the south Of course this can also be restated in more positive terms: nothing 
in the 'appendix' can throw serious doubt on a reconstruction that is based on 
a reading of the Islamic tradition, and that dates the fall of Khuzistan after 
that of al-Mada'in, perhaps in ah 22 or 23. 34 

What the Syriac account can tell us 

If the text cannot answer all of our questions, it can shed a direct and bright 

light on several others. It is to these questions that I shall now turn. 

The conquest of Jundaysabur 

The first problem concerns the fall of Jundaysabur. The sources familiar to 
al-Tabaii (wr 303/915) and al-Baladhuri (d. 279/892) held that the definitive 
conquest of Jundaysabur followed that of Tustar and al-Sus; this is the sequence 
that Donner describes. 55 But there were differing views: a tradition preserved 
by Khalifa ibn Khayyat (d. 240/854), for example, holds that Jundaysabur fell 
before Tustar, 56 and this is clearly what our Syriac authority has in mind as 
well Considering that the conquest of Jundaysabur does not seem to have 
been a principal concern for most of our Muslim authorities, and considering 
too that our Syriac source is not only local, but also that Jundaysabur was 
the metropolitan centre of Nestorian Bet Huzaye, 57 one might side with 
Khalifa In this case, as in others, consensus is apparently no guarantee ot 
accuracy. Meanwhile, what the Syriac source has to say about the canal- 
dominated topography of Tustar is very much in line with how the city is 
described in many conquest accounts in the Islamic tradition. 

The point to be emphasized here is a broader agreement between the 
Islamic tradition and our Syriac source: al-Sus and Tustar were among the 
last cities to hold out in Khuzistan, falling definitively only after Abu Musa 
appeared on the scene, and al-Hurmuzan, sent by Yazdagird, played a crucial 
role in the Sasanian defence. 

Al-Basra, al-Kufa, and the problem of conquest participation 

Our Syriac testimony on the founding of al-Basra and al-Kufa is one of the 
earliest datable accounts we possess. It is both familiar (the two are established 
as 'settlements' for the Arabs) and unfamiliar (Abu Musa al-Ash'an, rather 
than 'Utba ibn Ghazwan, being given credit for founding al-Basra). Another, 




54 Cf. Donner, Conquests, 217. ,, 

55 Al-Tabari, Tci'rTkh, I, 2567; al-Baladhuri, Futuh, 382; Donner, Conquests, 216. 

56 Khalifa ibn Khayyat, Tarikh, I, 138. 

57 For a detailed discussion, see Fiey, 'L Elam , 11 /-&/. i-jnu,™«. 
■ See, for example, al-Qumml (d. 805/1402^ Tankh-i Oumm, ^ r ^^S^vS^& 

Matba'at-i Majlis, 1934), 297— this work is a Persian translation of an a I "^fi^is P 
century Arabic original; see A. K S. Lambton,/An accoun of he Wnkh '^^%^± 
(1947-48) 586-96. Al-Qumml credits his material to Ibn Ishaq (d. 151/761) and Abu Ubaycia 
d 21 1/8^6) citing for the latter a Futuh ah! al-Islam, which seems otherwise unknown, the 
Sateri/lmay be familiar to Ibn al-Nadlm (w, 377/987) under the ■ mlc « g^ffifi55 
Ibn al-Nadlm, Fihrist, ed. Gustav Flugel (Leipzig: FC. W Vogel, 1871-72), tf IbnKhdllikan 
(d 681/1282) Wafayat al-a'van, ed. Ihsan 'Abbas (Beirut: Dar al-thaqafa, 1968-/2), V, >iJ. 
Whether this was an independent monograph, or rather a section in a arger work isat present 
hard to say— Michael Lecker thinks the former; see his 'Biographical notes on Abu UbayUa 
Ma'mar b. al-Mathanna', Studia Islamica 81 (1995) ,76. 

59 On al-Basra, see Charles Pellat, art. 'al-Basra' in EI 2 , I (Leiden: EJ.Bi ill, I960) lUS5d 
(which puis the conquest in ah 17); Salil? Ahmad al-'AU, ?- T ?Jtf?«^^ 
fil-Bwrafil-qarn ul-cm>wal al-hifrf (Baghdad: Malba'ai al-ma'anf, 1953), 25-6 (perhaps as early 
as ah 14 or 16). 




and admittedly much later, Christian source also credits Abu Musa with 
al-Basra, 60 but the evidence is more enticing than clinching. 

As faras the conquest is concerned, the Islamic tradition generally has 
Abu Musa al-Ash'an play a dual role. First, he is said to effect the definitive 
conquest of cities, such as Suq al-Ahwaz, that had reneged on earlier treaties; 
and second, he is given a prominent role in the two victories of al-Sus and 
Tustar, which broke the back of the Sasanian defence. As we have already 
seen, on the first of these our Syriac source can offer only silence, which is 
particularly frustrating since so many cities are said to have reneged on earlier 
agreements. In the case of Tustar we have another instance of this, but because 
our Syriac source does have something to say here, our conclusions perhaps 
have more force there. 61 On the second problem— Abu Musa's role in the 
Muslim armies— our Syriac source can suggest that credit for the conquest of 
al-Sus and Tustar indeed does belong to Abu Musa, rather than to other 
candidates favoured by our Muslim authorities, particularly Abu Sabra, whom 
Sayf ibn 'Umar (d. 180/796) gives pride of place in the army that besieged 
Tustar. 62 

It is not just the silence of our Syriac source that makes Abu Sabra's role 
at Tustar a problem. He is also curiously absent in the very battle scene that 
Sayf himself describes: it is at Abu Musa's feet, rather than Abu Sabra's, that 
the arrow shot from a traitor's bow dramatically lands, thus turning the tide 
of the battle. 63 It is true that his absence on the field could be argued away 
on the grounds that the conquest tradition occasionally distinguishes between 
a commander who has nominal authority over a campaign, and a sub- 
commander, sometimes called the amir al-qital, or 'battle commander', who 
leads the army into combat, and who has authority to enter into agreements 
on his superior's behalf. 64 But no such distinction is made at Tustar, and other 
sources are as consistent in ignoring Abu Sabra as they are on insisting on the 
command of Abu Musa. 65 

They ignore Abu Sabra's role in Tustar for the simple reason that they 
ignore him otherwise: Sayf is apparently alone in having him briefly hold the 
governorship of al-Basra after 'Utba ibn Ghazwan and before al-Mughlra ibn 
Shu'ba. 66 These then are the terms in which we can understand Abu Sabra's 
cameo appearance in Sayfs account, and the second reason why we should 
reject it. For it apparently comes not from an authentic memory of the events 
in question, but rather was generated by a view widely held by conquest 
authorises that the governorship of al-Basra and the leadership of the 
Khiizistan campaigns were one and the same. 67 In the case of Abu Musa, 

60 Bar Hebraeus (d. 1286), Ta'rTkh mukhtamr al-duwal, ed. Antoine SalhanT (Beirut: Imprimerie 
cathohque, 1890) 174, knows 'Utba and al-MughTra ibn Shu'ba only as military commanders- 
the laying out of the khitat, the building of mandzit and the congregational mosque, Arab 
settlement— all these are credited to Abu Musa. Al-Ya'qubT {Historiae, II, 163) explicitly credits 
Utba with the tkhtitat of the site. 

61 See below. ,. 4 '. 

f AI-TabarT, Ta'rTkh, I, 2553-6. ,.,; y . ,',_•, 

Al-labari, Ta'rTkh, I, 2553-6. 

63 ^'H 55 f-.. ..... 


^Thus Suhayl ibn 'Adi in ibid, I, 2506-7. 

'Thus tribesmen boast that they fought alongside Abu Musa; see Ibn AbT Shayba (d. 235/849) 
Al-Musanmp, ed. Sa'Td al-Lahham (Beirut: Dar al-fikr, 1989), VIII, 17; on p. 32 Abu Musa is 
explicitly identified as the amir al-jaysh). AI-QummT (TarTkh-i Oumm, 295), puts Abu 'Ubayda's 
and Ibn Ishaq s reports under the rubric dhikr-i fath-i Abu MSsa Ash'drl ;. ', ! i >->"■ i) I , ,i 
^ Al-Taban, r«'/f/c/;, I, 2498, 2550-51. ' ' ....... w ...., , ..' ., lJ: . £,j, . 

' Note that al-BaladhurT's first report (Futuh, 376: qahl), which outlines the : overall sequence 

of events, conspicuously and explicitly connects the conquest of al-Ahwaz withthe administration 

of al-Basra: They reported: al-Mughlra ibn Shu'ba raided Suq aI-Ahwa2 during his'governorship 

when Utba ibn Ghazwan was removed from al-Basra at the end of the yearlS-orthe-beginninR 

I m 1C y e . ar , 16 --- then Abu MuSt?i raided il when 'Umar appointed him governor of al-Basra after 

cii-ivi uglilra . .a < • i i" , . 



where we have a broad Islamic consensus that is corroborated by our Syriac 
source, there is good reason to think that the view is correct: Abu Musa 
founded al-Basra and did play a starring role in the conquest of Khuzistan. 
In the case of Abu Sabra we have only Sayf. 

Abu Sabra's obscurity may have had narrative advantages for Sayf, who 
has him oversee what is presented as two separate armies, one Basran and one 
Kufan. 68 These armies pose problems of their own. Now because our Syriac 
source implies that Abu Musa came to Khuzistan from al-Basra, we can put 
some stock in the Islamic accounts that speak of Basran armies as well. 69 
Kufan participation in the conquest of the south is altogether harder to confirm, 
however. As Donner has noted, 70 the introduction of reinforcements into the 
Khuzistan campaign — of which the Kufans under al-Nu'man ibn al-Muqarrin 
or 'Ammar ibn Yasir figure very prominently — was a matter of some 
controversy. In what follows I shall offer some suggestions why. 

The conquest of Khuzistan 

At issue was the region's revenues, since it was by claiming conquest experience 
that one argued one's share; in other words, the conquest record was influenced 
by post-conquest politics. 71 Sayf preserves an account that has al-Ahnaf ibn 
Qays voicing Basran grievances vis-a-vis the Kufans soon after the conquest 
of Suq al-Ahwaz, and to judge by 'Ulnar's response, his argument was convin- 
cing: in addition to doling out to the Basrans former Sasanian crown land, 
'Umar is said to have increased the number of Basrans receiving 2,000 dirhams 
by including among them all those who had fought at (Suq) al-Ahwaz. 72 The 
Basrans and Kufans disputed about Tustar in particular. The categorical 
assertion that Tustar belongs to the Basrans is warning enough that adminis- 
trative geography was controversial, 73 and echoes of the controversy can be 
heard even as late as Yaqut's time, when some apparently claimed that Tustar 
belonged to al-Ahwaz, while others held that it belonged to al-Basra. Yaqiit 
also tells of a heated exchange between the two parties that took place before 
'Umar, each claiming Tustar as their own. 74 Ibn A'tham al-Kufi has a much 
longer version of this, or a similar, scene. 

The Basrans and Kufans came to argue, the Basrans saying: The conquest 
is ours!' and the Kufans saying: 'No, the conquest is ours!' So they argued 
about it to the point that something truly disagreeable almost happened 
between them. 75 

68 Al-Tabari, Ta'nkh, I, 2553: wa-'ala l-fariqayn jamTan Abu Sabra. Cf. al-Baladhuri, Fatah, 
380: hamila ahl al-Basra wa-ahl al-Kufa. Cf. the much latercase of al-Muhallab and 'Attab ibn 
Warqa', where the position of amir al-jamaa ( = amTr al-qital) is determined by conquest claims 
by Basrans and Kufans; see al-Mubarrad (d. 285/898), Al-Kainil JT l-lugha wa-l-adab, ed. William 
Wright (Leipzig: F. A. Brockhaus, 1874-92), 675. 

^Al-Tabari, Ta'nkh, I, 2552-54; Ibn A'tham, Kitab al-futah, ed. Muhammad 'Abd 
al-Mu'Id 'Khan et al. (Hyderabad: Da'irat al-ma'arif al-'uthmanlya', 1388-95/1968-75), II, 5; 
al-Baladhuri, Fatah, 372-3. 

70 See Donner, 'Conquests, 342 n. 229. 

71 Cf. Robert Brunschvig, 'Ibn 'Abdalhakam et la conquete de l'Afrique du Nord par les 
arabes: etude critique', AlEO 6 (1942-47), 'l 10-55. Cf. the case of the Jazira in C. F. Robinson, 
Empire anil elites after the Muslim conquest: the transformation of northern Mesopotamia 
(Cambridge: Cambridge University Press, 2000), 6ff. 

72 Al-Tabari, Ta'nkh, I, 2539-40; cf. I, 2672-3. 

73 Cf. Ibn AbT Shayba, Musannaf, VIII, 35: wa-Tustar min aril al-Basra. 

74 Yaqut, Mu'jam al-buliidii, I, 849 (both accounts). 

75 Ibn A'tham, Futuh, II, 27. Cf. the dispute between a Kufan and a Syrian, where the former 
crows about his townsmen's victories: 'We were the victors at the battle of al-Qadisiya and the 
battle of such-and-such' (nahnu asltab yawn al-Qadisiya wa-yawm kadhci wu-kddlid...), and the 
latter about his townsmen's victories (including al-Yarmuk), in Ibn AbT Shayba, Musannaf, 
VIII, 17. 



The solution that 'Umar is here given to provide holds that although the 
conquest is indeed to be credited to the Basrans, its benefits accrue to Basrans 
and Kufans alike. 76 

'Umar's view that conquest revenues were to be distributed to the Basrans 
and Kufans is more fully described by Ibn A'tham. Here Abu Musa al-Ash'an 
writes to the caliph, requesting reinforcements for the upcoming battle at 
Tustar; the caliph responds by dispatching a Kufan commander. 'Ammar ibn 
Yasir. As in other reports, 77 the operative terms (istamadda, amadda) are 
topological, in this case probably employed not only to emphasize the role of 
the caliph in conquest decision making, 78 but also to bring Kufan troops into 
a picture that had been dominated by Basrans. 'Ammar ibn Yasir is then given 
to describe the contents of the letter from 'Umar: 'He (the caliph) is ordering 
me to march to Abu Musa al-Ash'arl to come to the aid of our believing 
brethren from al-Basra' {li-nusrat ikhwanina al-mu'mimn min ahl al-Basra). 19 
Then, after the battle, 'Umar passes judgement on the ensuing controversy: 80 

Tustar is [to be considered] among the conquests (maghazT) of the Basrans 
even though they were aided by their brethren from among the Kufans 
{innama nusiru bi-ikhwanihim min ahl al-Kufa). The same thing goes for the 
Kufans: if they make raids in their marches (thughur), and the Basrans 
come to their aid, there is no harm [done to their claim] (lam yakun 
bi-dhalika bas). For according to the book of God, victory belongs to [all] 
the believers; God has made [all] the believers brethren. 81 The conquest is 
the Basrans', but the Kufans are their equals in the rewards and spoils 
(shuraka'uhum fi l-ajr wa-l-ghamma). Beware the discords inspired by 
Satan! 82 

A post-conquest opinion on the division of spoils — i.e. that merely by assisting 
(misva) the Basrans, the Kufans had earned a full share — is thus detectable in 
a tradition that purports to describe the conquest itself. That precisely this 
issue was controversial is made clear elsewhere, in a work that is explicitly 
legal in character. 83 The late and polemical character of the account explains 
'Umar's eirenic tone: all the rivalry that we might expect of campaigning 
armies, and of which we have clear echoes in the post-conquest disputes, 84 is 
stifled by a unitary and providential view of conquest history. 

Post-conquest disputes influenced the historical record in other ways as 
well. If some attributed to 'Umar the view that the Basrans and Kufans were 
to share equally in the spoils, others thought differently. Thus Yaqiit preserves 
an echo of another view, which held that 'Umar granted the revenues of Tustar 
to the Basrans rather than to the Kufans, on the grounds that it was closer to 
al-Basra than it was to Kfifa. 85 In one of the titles attributed to al-Mada'inl 

76 Ibn A'tham, Futiih, II, 27. 

77 Al-Tabari, Ta'rJkh, I, 2534. 

78 See Noth/Conrad, Early Arabic historical tradition, 123-6. But cf. C. F. Robinson, The 
study of Islamic historiography: a progress report', JRAS 3, 7, 9 (1997), 218 ff. 

75 Ibn A'tham, Futuh, II, 10. 
B0 rbid,\l,21. 

81 Cf. Surat al-Anfal (8), v. 74; Surat al-Rum (30), v. 47; Surat al-Hujurat (49), v. 10. 

82 The vocabulary remains quranic: see, in particular, Surat al-Nisa' (4), v. 12; Surat Yusuf 
(12) v. 100; Surat al-Rum (30), v. 28. 

83 See al-Taban, Ikhtilaf al-fuqaha\ ed. Joseph Schacht (Leiden: E. J. Brill, 1933), 68-71. 

8A One can only wonder about the contents of the Fakhr ahl al-Kufa 'aid l-Basrd by al-Waqidl 
(d. 207/823), and the Mufdkharat ahl al-Basra wa-ahl al-Kufa by al-Mada'inl (d. 228/842); on 
which se? Ibn al-Nattim, Fihmi< IQ9< 104. Cf. $f§& aLYa'qSbi, KHab aUhuldmi, ed. M. J. de Ooefe 
(Leiden: E. J. Brill, 1885; BGA 7), 161-13. 

85 Yaqiit, Mu'jam al-buldati, I, 849: fa-ja'alaha 'Umar ibn al-Kliattab min ard al-Basra 
li-qurbiha minha. 



(d. 228/842), the Khabar al-Basra wa-futiihiha wa-futfih ma yuqaribuha min 
Dahistan wa-l-Ahwaz wa-Masabdhan wa-ghayr dhalika, 86 we may have a 
reconstruction of conquest history according to this principle. 

Hinds has shown how Basran participation in the initial conquest of Fars 
could be exaggerated by our sources. 87 Given the problems surrounding the 
Kufans in Khuzistan, those determined to reconstruct history could do worse 
than to rethink the Kufans' role here. 

The question of treaties 

Things are perhaps only slightly less thorny when it comes to what our Syriac 
source calls a 'truce' (shayna). That the campaigns in Khuzistan were inter- 
rupted by a short-lived peace is clear enough; the problem is that the one 
promising account we have in the Islamic tradition, which is Sayf's, identifies 
al-Hurmuzan and 'Utba ibn Ghazwan, rather than al-Hurmuzan and Abu 
Musa al-Ash'arl, as the parties concerned. 88 

In fact, Sayf knew of two such agreements. Al-Hurmuzan is first said to 
have reached a sulh agreement with 'Utba at Sviq al-Ahwaz, after he had heard 
of the losses of Manadhir and Nahr TIra to Muslim forces: 

When the [Muslim] fighting force (al-qowm) moved against al-Hurmuzan 
and encamped near him in al-Ahwaz, he saw that he lacked the force to 
do battle. So he requested a sulh. They (the Muslims) then wrote to 'Utba 
about the matter, requesting his instructions. Al-Hurmuzan wrote to him, 
and 'Utba agreed to the offer on the following terms: [al-Hurmuzan would 
retain] all of al-Ahwaz and Mihrajanqadhaq, except Nahr TIra, Manadhir, 
and that part of Siiq al-Ahwaz that they (the Muslims) had overrun. What 
we have liberated will not be returned to them. 

A dispute is then said to have arisen concerning the borders between 
al-Hurmuzan's territory and that of the Muslims; in the aftermath, al-Hurmuzan 
'reneged (kafaro), withheld what he had accepted, 89 enrolled Kurds (in this 
army), and so his army grew strong'. 90 He then took to the field, was defeated 
at Suq al-Ahwaz, and eventually fled to Ramhurmuz. There he reached a 
second sulh, and once again 'Umar is given to impose conditions: "Umar 
ordered him ('Utba) to accept [al-Hurmuzan's offer], on the following terms: 
that the land not conquered, i.e. Tustar, al-Sus, Jundaysabur, al-Bunyan, and 
Mihrajanqadhaq [would come under Muslim authority]'. Al-Hurmuzan agreed 
to the terms, which are now described in more detail: 

The commanders of the Ahwaz campaign took responsibility for what was 
assigned to them, and al-Hurmuzan for his sulh, [the latter] levying taxes 
for them, and [the former] protecting him. 91 If the Kurds of Fars raided 
him, they would come to his aid and defend him. 92 



86 See YaqQt, Irsliad al-anb ila mu'rifut al-mtib, ed. D. S. Margoliouth, 2nd ed. (Leiden: 
E. J. Brill, and London: Luzac, 1923-31), V, 315; Ursula Sezgin, Abu Milmuf. Ein Beit rag zur 
Historiographs tier umaiyadischen Zeit (Leiden: E. J. Brill, 1971), 44. 

87 Martin Hinds, The first Arab conquests of Fars', Iran 22 (1984), 39-53; reprinted in his 
Studies in early Islamic history, ed. Jere Bacharach, Lawrence I. Conrad and Patricia Crone 
(Princeton: Darwin Press, 1996), 197-229. 

88 Al-TabarT, Tu'rikh, I, 2538-42. 

89 i.e. what he had agreed to yield in tribute? The Arabic text is wa-mana'a ma qahihihu. 

90 Al-Jabarl, Ta'rikh, I, 2540. 

91 yamna'unahit; one might also read yu'awinuhwn, 'and he (al-Hurmuzan) offering aid to them'. 

92 Al-Tabari, Tarlkh, I, 2543. 



This sulh fared no better than the first: after two Muslim forces were sent into 
al-Ahwaz, one of which was led by Abu Musa, al-Hurmuzan engaged 
al-Nu'man ibn al-Muqarrin, was defeated, and fled to Tustar. 93 

Once again, one can be cheered by the common ground: al-Hurmuzan 
seems to have entered into some kind of agreement with the Muslims, which 
perhaps stipulated an exchange of tribute for recognition of local authority, 
and during which al-Hurmuzan reinforced his armies. Although its exact timing 
escapes us, it must have been reached during, or soon after, the fall of al-Ahwaz. 
But it is difficult to say much more. The close similarities between SayPs two 
agreements might be taken to suggest either that the 'appendix 1 conflated the 
two, or that Sayf (or his sources) had so heavily elaborated a single truce 
account that out of its precipitate emerged two separate accounts. A tentative 
argument, might be made in favour of SayPs second treaty. For whereas the 
first says nothing explicit about tribute, the second clearly stipulates that 
al-Hurmuzan collect taxes for the Muslims. Moreover, it is only at this point 
that Abu Musa enters the scene, and it is here too that Tustar emerges as a 
stronghold for al-Hurmuzan: to Tustar he withdraws after his defeat, and to 
Tustar comes help from the people of Fars. Finally, a later passage that 
mentions 'the rebellion (intiqad) of al-Hurmuzan' clearly alludes to the breaking 
of the second treaty. 94 

Al-Sus: leadership, Asawira, and Daniel 

Since our Syriac source places Hormizdan at both Shush and Shushtra, and 
describes his capture in the latter, we are to infer that it fell after Shush. 
Donner argues the opposite, putting al-Sus after Tustar. 95 On this sequence 
no authority is cited, but it is implicit in Sayf in al-Taban, 96 and explicit in 
al-DInawarl (d. 282/891 ). 97 There appears to have been some disagreement on 
the matter, however. Al-TabarT freely volunteers that there was no consensus 
about the conquest of al-Sus, 98 al-Baladhuri discusses Tustar after al-Sus, 99 
and Ibn al-A'tham, as well as Abu 'Ubayda (d. 211/826) and Ibn Ishaq 
(d. 151/761, as preserved by al-QummT, d. 805/1 402),. clearly put the fall of 
al-Sus before that of Tustar. 100 This was Caetani's view, 101 and it is vindicated 
by our Syriac source. 

In the precise course of the conquest of al-Sus the Islamic sources evince 
little interest. A failed ruse attempted by al-Sus's (anonymous) marzban is 
featured in one of al-Baladhuif s accounts, according to which an aman was 
granted, and where there is no suggestion that the city was penetrated;' the 
point is that Abu Musa saw through the marzbaii's trick, executing him and 
80 fighters {muqatild) as a result. 102 A version of the Same story is then related 
by a participant in the battle; here we read of an anonymous dihqan. 103 Ibn 

93 ibid, 1,2552-3. 

94 ibid, 1,2614. 

95 Donner, Conquests, 216. 

96 Al-TabarT, Ta'rTkh, I, 2551-6. 

97 Al-DInawarl, Akhbar, 140. 

98 Al-TabarT, Ta'rTkh, I, 2561: ikhlalafa ah! al-siyar fi amriha. Sayfs account of the conquest 
of Tustar {ibid., I, 2542-5) may be out of place. 

m AI-BaladhurT, Futilh, 378-8 1 . 

100 Ibn A'tham, Futuh, II, 9: thumma sara Abu Musa ila Tustar ba'dfaraghihi mitt amr al-Sus; 
al-QummT, Tarikh-i Qim'mi, 295 (al-Sus follows Manadhir). i 

l01 Caetani, Annali dell'Islam, IV, 454. 

102 Al-Baladhuri, Futilh, 378. 

103 ibid, 378-9; see also Ibn AbT Shayba, Musannaf, VIII, 32; al-QummT, Tarikh-i Qumm, 295. 



A'tham has a version of the same story, but now both the marzban (Sabur ibn 
Adharmahan) and a lieutenant are given names. 104 Meanwhile, Sayf seems to 
be at pains to demonstrate the clemency of the victorious Muslims: the city is 
stormed after a siege, the conquered pathetically beg for mercy, and a sulk is 
granted by the Muslims, who are apparently led by Abu Sabra, although Abu 
Musa is also present. Behind the tradition — and perhaps the trickery account 
as well — there are signs of some disagreement: the granting of a sulh after the 
Muslims' violent entrance (bad ma dakhaluha 'anwotan), and the division of 
spoils that is said to have taken place before the sulh (wa'qtammu ma asabil 
qabla l-sulh) suggest that this is a reconciling account, 105 intended to accom- 
modate conflicting sulh and 'anwa traditions. 106 The failed ruse may perform 
a similar function for Ibn A'tham: spoils were taken after an aman because of 
the trickery. 107 Certainly our Syriac account, which details the killing of 
Christians in the city, does not inspire much confidence in reports such as 
these. In none of these Islamic accounts does al-Hurmuzan appear. 

In Sayf s report al-Hurmuzan is again absent in the Sasanian defence, but 
we may have an echo of his presence: al-Shahriyar, said to be al-Hurmuzan's 
brother, leads the Muslims in battle. It is here that we get a glimpse at what 
really concerned the authorities: the fate of the asawira, the elite cavalry of 
the Sasanian army. The asawira, like so much in early Islamic history, are only 
now beginning to receive their due, and although the conquest accounts have 
generally been enough to persuade historians that they converted in this 
period, 108 there is some evidence to suggest that their conversion is a product 
of the Umayyad period. 109 For early Muslim traditionists it was probably not 
so much their conversion that was at issue as the top stipends that they were 
awarded; that al-Baladhurl devoted an entire section to aim al-asawira wa-l-zuff 
at least suggests that the issue retained some interest as late as his day. 110 On 
the one hand, there was a view that the asawira remained loyal to the Sasanians 
through Tustar. Thus Ibn A'tham, whose sequence follows that of our 
Syriac source, has no problem in putting not only maraziba, but also asawira 
in al-Hurmuzan's forces that resisted the Muslims at Tustar; 111 Ibn Sa'd 
(d. 230/844) also preserves a reconstruction of events that has al-Hurmuzan 
commanding a group of asawira at Tustar. 112 On the other hand, al-Mada'inl 
seems to reflect a widely held view that Siyah al-Uswarl was sent by Yazdagird 
to defend al-Sus, while al-Hurmuzan was sent to Tustar; and when, according 
to al-Baladhuri's sources, Siyah learned of the capitulation of al-Sus, or, 
according to al-Mada'inl, came to realize more generally that the Muslims 

> ■ 

104 Ibn A'tham, Futuh, II, 6-7. 

105 Cf. the case of Sfiq al-Ahwaz, about which Khalifa ibn Khayyat (Ta'rJkh, I, 106) reports 
that it was conquered sulhan eiw 'anwatan. 

106 Al-Tabari, Ta'rfkh', I, 2565. Cf. Albrechl Noth, 'Zum Verhaltnis von Kalifer Zentralgewalt 
und Provinzen in umayyadischer Zeit. Die 'Sulh-'Anwa' Traditionen fur Agypten und den Iraq', 
Wl 14(1973), 150-62. 

107 Ibn A'tham, Futuh, II, 7. 

108 Bertold Spuler, Iran im friihishimischer Zeit (Wiesbaden: Franz Sleiner, 1952), 254; Morony, 
Iraq, 198. On the asawira in early Islam in general, Mohsen Zakeri, Sdsanid soldiers and 
early Muslim soeiety: the origins of 'Ayyarun and Futuwwa (Wiesbaden: Harrassowitz, 1995), 
index, s.v. 

Patricia Crone, Slaves on horses: the evolution of (he Islamic polity (Cambridge: Cambridge 
University Press, 1980), 237 n. 362. 

110 Al-Baladhurl, Futuh, 372-3. Among the titles attributed to al-Mada'inl is a Kitah al-asdwiru; 
see Rida Tajaddud's edition of the Fihrist (Tehran: Matba*at-i Danishgah, 1971), 115; Bayard 
Dodge, The Fihrist ofal-NadTm: a tenth-century survey of Muslim culture (New York and London: 
Columbia University Press, 1970), I, 225 (Flugel, Fihrist, 103, reads Kitah al-ishara). 

111 Ibn A'tham, Futuh, II, 13. 

112 Ibn Sa'd, Kitah al-tuhaijdt al-kablr, ed. Eduard Sachau et at. (Leiden: E. J. Brill, 1904-40), 
V, 64. 

, _ TT ^,.-.,„-,— ,..,.- . T: - ;j -;—^^£totjtu^ "*•■"■.-' wstszira 

NBMIPifml ^^^ 



were overwhelming the Sasanians, he and the asawiro enrolled in the Muslim 
armies instead. 113 This opens the door — perhaps only narrowly — for the 
participation of the asawira at Tustar, which was reluctantly conceded. 114 

Indeed, it is only by presuming that they converted before Tustar that we 
can understand Sayf s version of events. For Sayf has it that 'Umar ordered 
Abu Musa to assign them the highest stipend, equal to that granted to any 
Arab tribesman, even though Abu Musa had nothing but disdain for their 
feeble effort at Tustar. A few lines of poetry that follow give voice to consequent 
Arab resentment: 

When 'Umar (al-faniq) saw the excellence of their valor 

And came to see what might come of the matter, 115 

He assigned to them a stipend of two thousand, 

Having seen fit to give the 'Akk and Himyar a stipend of three hundred. 116 

Reports that identify STnah/STneh as the traitor who betrayed Tustar to the 
Muslims presumably reflect the same nnti-asawira sentiments that produced 
these lines. 117 

We are on firmer ground concerning Daniel. The legendary connection 
between Daniel and al-Sus is not an Islamic invention. 118 It had been made 
before Islam, 119 and by the seventh century (if not earlier) it appears to 
have gained wide currency. Thus, the Armenian history attributed to Sebeos 
(wr. c. 660-70) relates that the Byzantine emperor Maurice (r. 582-602) 
made an unsuccessful attempt to remove Daniel's body from al-Sus to 
Constantinople; as in our Syriac account, here too various claims were made 
about the identity of the deceased. 120 It is in the light of this material that we 
should read our Syriac account: 'they [the Arabs] seized the house that is 
called the "House of Mar Daniel", and took the treasure there enclosed, 
which had been kept there on the kings' orders since the days of Darius and 
Cyrus'. It is in the same light that we should also read the Arabic accounts of 
how Daniel's body was discovered in al-Sus; these are positively ubiquitous in 
the conquest tradition. 121 

As late antique monotheists, the conquering Muslims might be expected to 
have taken an interest in Daniel, in this period considered a prophet not only 
by Christians, but also by some Jews. 122 He does not appear in the Quran, 
but remembering that this inventory was not complete, 123 and assuming as 

113 Al-TabarT, Ta'nkh, I, 2562-4; al-Baladhuri, Futuli, 372-3, on the authority of 'a group of 
learned men' {jama a mitt ahl al-'ilm). 

114 AI-BaladhurT, Futilh, 3S2:yuqal...)va- Allah a'lam, 'it is said. ..but God knows best', al-Tabarl, 
Ta'nkh, I, 2563, 2564: wa-qawm yaqilhlna, 'there are some who say'. 

u5 i.e. he recognized their potential, as well as the hazards of putting them ofF. 

116 Al-TabarT, Ta'nkh, I, 2563-4. 

117 On the betrayal of Tustar, below. 

118 Which seems to be implied by William Brinner in his translation, The history of al-Tabari, 
II: Prophets and patriarchs (Albany: State University of New York Press, 1987), 48 n. 129; and 
Georges Vajda, art. 'DaniyaT in EI 1 , II (Leiden: E. J. Brill, 1965), 1 12b. 

1,9 See the evidence gathered by Louis Ginzberg in his The legends of the Jews (Philadelphia: 
Jewish Publication Society of America, 1928), VI, 437 n. 20. 

120 Histoire d'Heraclius, 29-30. Cf. al-BaladhurT, Futilh, 378: qila inna fill i jut h that Daniyal. 

121 Al-TabarT, Ta'nkh, I, 2566-7; al-Baladhuri, Futil/i, 378; Ibn A'tham, Futilh, II, 6-9; Ibn 
AbT Shayba, Musannaf, VIII, 31; al-Qummf, Tankh-i Qiimm, 296-7; Ibn Zanjawayh (d. 251/865), 
Kitab al-amwal, ed. Shakir Dhlb Fayyad (Riyadh: Markaz al-Malik Faysal li-1-buhuth wa-1-dirasat, 
1986), II, 748; Ibn AbT 'Adasa (fl. 9th/i5th c), Qisas al-anbiya\ KhalidI Library (Jerusalem), Ms. 
Ar. 86, fol. 1 14r. See also M. Kevran and S. Rehimel, 'Suse islamique: retnarques preliminaires 
et perspectives', Stadia Iranica 3 (1974), 256. 

122 See John Barton, Oracles of God: perceptions of prophecy in Israel after the exile (London: 
Darton, Longman and Todd, 1986), 36-7, 99; Geza Vermes, 'Josephus' treatment of the Book of 
Darnel', JJS, 42 (1991), 158 with n. 14. 

123 See Surat Ghafir (40), v. 78: wa-la-qad arsalna rusulan min qablika minhum man qasasna 
'alayka wa-minhum man lam naqsus 'alayka, 'We sent Messengers before thee; of some We have 





well that this inventory gradually created, rather than reflected, a consensus, 
one might speculate that the conquering Muslims had open minds. 124 

In contrast to the attitudes of the conquering Muslims, the concerns of the 
later traditionists are fairly clear. First, Daniel's prophecies enjoyed some 
popularity in the early period, and this almost certainly reflects the broad 
appeal of apocalyptic texts among Christians 125 and Muslims alike. 126 In fact, 
Sayf (or one of his sources) betrays an Islamic triumphalism that is only fully 
intelligible in the light of Christian millenarian anxieties that tied the conquest 
of al-Sus to the eschaton. Sayf reports that the monks and priests (al-ruhban 
wa-l-qassTsun) mocked the besieging Muslims from the top of the walls of the 
city: 'O host of Arabs, among the things taught us by our learned men and 
ancestors is that only the Antichrist, or an army led by the Antichrist (qawm 
fihim al-dajjal), will conquer al-Sus. If the Antichrist is leading you, you will 
take it (al-Sus); if he is not, don't bother besieging us'. 127 Of course in the 
eyes of Muslim informants the conquests were the work not of the Antichrist, 
but of God Himself; and far from marking the beginning of the End, they 
came to mark an altogether new beginning. The successful siege of al-Sus thus 
makes a mockery of the Christians and their misplaced trust, turning what 
must have been a familiar topos on its head. 128 

The Daniel tradition seems to have been informed by iconoclastic concerns 
as well. 129 Here it may be significant that the Syriac does not corroborate 
the Islamic accounts that describe the Arabs' relocation of Daniel's body. 
Although the story is recounted in several different ways, 130 all are drawn 
together by a shared concern to make the site inaccessible to those determined 
to locate — and perhaps translate — relics. 131 

related to thee, and some We have not related to thee' (Arberry). CL the relatively early discussion 
in 'Abd al-Malik ibn Habib (d. 238/852), Kitab al- ta'rikh, ed. Jorge Aguade (Madrid: Consejo 
superior de investigaciones cientificas, 1991), 26-7. 

124 In fact, occasional passages in the Islamic sources echo the Rabbis' rejection of his prophetic 
status, and sound like special pleading. Note, for instance, the words attributed to Abu Sabra 
(al-Taban, Ta'rikh, I, 2566), but particularly those of 'AIT, who answered a query by stating: bald 
liddhd Daniyal al-hakim wa-huwa ghayr mursal (Ibn A'tham, Futuh, II, 8); cf. Ibn Abi Shayba, 
Musannaf, VIII, 3*1; fa-innahu nabT (but not, it appears, a rasul). For a particularly rich discussion 
of rasfil and nabT, see Geo Widengren, Muhammad, the Apostle of God, and his Ascension (King 
and Saviour V) (Uppsala: Almqvist and Wiksells, 1955), chapters 1-4. 

125 For the use of the Danielic paradigm in apocalypses and histories, see G. J. Reinink, 
'Ps.-Methodius: a concept of history in response to the rise of Islam', in Averil Cameron and 
Lawrence I. Conrad, eds, The Byzantine and early Islamic Near East, I: problems in the literary 
source material (Princeton: Darwin Press, 1992), 161—6; and (in the same volume), H. J. W. Drijvers, 
'The Gospel of the Twelve' Apostles: a Syriac apocalypse from the early Islamic period', 201-08. 

126 See al-Khatlb al-Baghdadl (d. 463/1071), Taqyld al-'ilm, ed. Yiisuf al-'Ushsh (Damascus: 
Dar ihya' al-sunna al-nabawiya, 1949), 51, 56-7 (a scribe from al-Sus copies the Book of Daniel 
and is scolded for doing so; first noted by Crone, Slaves, 18). On the popularity of Daniel among 
early Sasanian Jews, see Jacob Neusner, A history of the Jews in Babylonia, II: The early Sasanian 
period (Leiden: E. J. Brill, 1966), 236-7. 

127 Al-TabarT, Ta'rikh, I, 2564-5. 

128 The presence of the Antichrist in a besieging army has a long tradition in Christian writing; 
for a fourth-century example, see Norman Colin, The pursuit of the millennium, revised ed. 
(London: Pimlico, 1993), 27-8. 

129 A strong aversion to relics and icons is attested in an early eighth-century source from 
southern Iraq; for a brief summary of the unpublished Syriac disputation between a monk of Bet 
Hale and an Arab, see G. J. Reinink, trans., Die syrische Apokalypse des pseudo-Methodius 
(Louvain: Peeters, 1993; CSCO 541, Scr. syri 221), xlviii. See also Hoyland, Seeing Islam as others 
saw it, 465-72; for some tentative archaeological evidence for Islamic iconoclasm, see Robert 
Schick, The Christian communities of Palestine from Byzantine to Islamic rule: a historical and 
archeological study (Princeton: Darwin Press, 1995), 207-09. 

130 Al-Baladhuri, Fundi, 378; al-Taban, Ta'rikh, 1, 2567; Ibn-A'tham, Futuh, II, 8-9. 

131 The reason is made explicit by Ibn A'tham (Futuh, II, 8), who has 'All recommend that the 
body be reburied 'in a place where the people of al-Sus would not be able to find his grave', cf. 
Ibn AbT Shayba, Musannaf, VIII, 31-2, on a Tustar corpse discussed below: 'a place known only 
to you two'. According to al-Qumml (TarTkh-i Qumm, 297), only some QummTs who just happened 
to be in al-Sus were told of its location. 



Tustar I: traitors and treaties 

If the historiography of the conquest of Khuzistan has generally moved little 
from Wellhauserfs day, an exception is the siege of Tustar. 132 The historicity 
of this siege was accepted by Wellhausen and Caetani, 133 and continues to be 
accepted elsewhere; in some quarters this also includes an act of treachery on 
the part of a Tustar! local, which delivered the city into the Muslims' hands. 134 
But with Noth we finally have a dissident voice. Pointing to the multiplicity 
of siege accounts in the Islamic conquest traditions in general, and adducing 
the Tustar account in particular, he argues that they must be interpreted as a 
feature of historical discourse: they represent 'not the reporting of history, but 
rather the deployment of literary stereotypes'. 135 

In general terms, Noth is certainly correct: siege/betrayal accounts can 
function stereotypically, 136 'drifting' from one event to the next. 137 It may be 
that the appearance of the iopos in the futuh literature is in some way related 
to the treacherous Jew of the sTra. 138 Since the repertoire of pre-Islamic Syriac 
historical writing includes siege accounts of great drama, 139 one might also 
suggest that it was popular enough to circulate widely in the Near East of late 
antiquity. 140 In any case, just as a specific takbTr account can be corroborated 
by an early Syriac source, 141 so too, it appears, can the occasional siege. In this 
particular case, accounts that relate a siege and betrayal quite clearly reflect 
an early — and authentic — memory of events. For there is Syriac corroboration 
not only for the betrayal of the city, but also for the length of the siege (two 
years), 142 as well as for the Muslims' penetration of the city through water 
tunnels under its walls. 143 

I3 -The siege is very well attested in the Islamic sources; see Khalifa ibn Khayyat, Ta'rTkh, I, 
133. 138-42; Ibn AbTShayba, Musannaf. VIII, 28-32; al-Tabarl, Ta'nkh, I, 2552-6; al-Baladhuri, 
Fulfill, 380; al-DInawaff, Akhbar, 137-8; Ibn A'tham, Futilh, II, 12-15, 18-23; Ibn Sa'd, Tabaqat, 
V, 64- al-Qumml, TarTkh-i Qumm, 297-8. See also Bar Hebraeus, Ta'rTkh, 174. 

'•"Wellhausen, 'Prolegomena', 96; Caetani, Annali dell 'Islam, IV, 457-8. 

134 D. R. Hill, for example, considers: That the entry was effected through the treachery of a 
citizen is quite probable, the Muslims at this time being ineffectual in siege warfare'. See his The 

Shustar was protracted, but in the end an Iranian's treachery— his name was Siya— enabled the 
Arabs to enter the city'. 

135 Noth/Conrad, Early Arabic historical tradition, 19. 

136 The traitor topos is also noted by Lawrence I. Conrad, 'The conquest of Arwad: a source- 
critical study in the historiography of the early medieval Near East', in Cameron and Conrad, 
eds, The Byzantine and Early Islamic Near East, I, 363. 

137 On : drift\ see Noth/Conrad, Early Arabic historical tradition, 109. : 

138 Conrad, 'Arwad', 363. citing Wansbrough, Sectarian milieu, 18-21, 109 (on the motif of 
the treacherous Jew in the sTra tradition). 

139 Of the many examples that could be cited, see ps.-Zacharias Rhetor (wr. c. 550),. Historia 
ecclesiastica Zachariae Rhctori vulgo adscripta, ed. and trans. E. W. Brooks (Paris: L, Diirbecq, 
1919-21; CSCO 83-84, Scr. syri 38-39), Vll.iii-iv (25-28/16-19), IX.xvii (132-33/90-91). 
Similarly, The Chronicle of Joshua the Stylite (written c. 518), ed. and trans. William Wright 
(Cambridge: Cambridge University Press, 1882), 50/42 (guards fall asleep): 'Whether then through 
this remissness, as we think, or by an act of treachery, as people said, or as a chastisement from 
God...' (Wright's translation); and 68/59 ('deserter' helps Byzantines against Persians). Also, 
compare the final section of translated Syriac above (a Qatar! colludes with someone who has a 
house on the city walls) with ps.-Joshua, 69/59-60 (defenders have built temporary houses on the 
walls); are we to understand that the co-conspirator was part of the force defending the city? 

! *'° It almost goes without saying that stories such as these have a very long tradition. Cf. 
Joshua 2, which describes how Rahab, a harlot in Jericho, admits, shelters, and cuts a deal with 
Israelite spies that guarantees the safety of her family; for a discussion and bibliography, see 
J. Alberto Soggin. Joshua: a commentary (London: SCM Press, 1972), 34-43. 

M ; Crone, Slaves, 12. • ; .. i 

142 Thus Ibn Sa'd, Tabaqat, V, 64 (a variant also proposes eighteen months); Khalifa ibn 
Khayyat, Ta'rTkh, I, 139 (around a year), 141 (two years or eighteen months); IbnAbT Shayba, 
Musannaf, VIII, 28 (around a year). < . 

143 Thus Ibn AbT Shayba, Musannaf, VIII, 28: fa-adkhalahu mill madkhal al-mci '; Khalifa ibn 
Khayyat, Ta'nkh, I, 139: fa-adkhalahu min madkhal al-ma' madkhalan; Ibn A'tham, Futilh, II, 20: 



Of course this is not to say that we should accept the Tustar traditions in 
their entirety. For sieges produce tales: tales of courage, piety, steadfastness, 
of clemency, arrogance, and hubris. As Noldeke remarked, 144 the particularly 
long seige of Tustar produced its share of stories, and these probably explain 
why the conquest was invoked in apparently stereotypical fashion. 145 We may 
even have a very brief glimpse of the Sitz im Leben of some of the storytelling. 
Asked by 'Umar to speak on the conquest of Tustar, 'Ziyad (ibn Abihi) arose 
and spoke with such skill that the people were astonished by his eloquence, 
proclaiming: Ibn 'Ubayd is a khatTbV 146 Needless to say, a performance such 
as this one earned praise not for its dogged fidelity to what happened, but by 
moving people; what mattered was not a close correspondence to historical 
truth, but rather the speaker's impressive command of a rhetoric that told a 
great story. Since the process by which memory was clouded by tale-telling 
was already well under way when we get our first look at our traditions, there 
is no question of finding an Islamic account that has survived unaffected: 
legendary material crowds our early accounts (Ibn Sa'd, Ibn Abi Shayba, and 
Khalifa ibn Khayyat), 147 as it crowds our later sources. 

Now some of this material, such as the legendary awa'il, we can safely argue 
away, not only because they are usually so transparent, but also because they 
are often expendable: no serious interpretation of the conquest of Tustar turns 
on 'the first to light the fire at the gate of Tustar'. 148 The point I would emphasize 
here is the difficulty of distinguishing between the baby and the bath. Without 
our Syriac text, for example, we would not know that it was apparently only 
the identity of the traitor that was conditioned by polemics. In most of the early 
accounts the traitor remains stubbornly anonymous, 149 but exceptions are 
al-DInawari and Abu 'Ubayda/Ibn Ishaq (as preserved in al-Qummi); in both 
cases the figure starts out anonymously (rajul mm ashrafahl al-madma, dihqam 
az jumleh-i buzurgan-i Tustar), but is then identified as a certain STna/Sineh 
(wa'smuhu Sma, nam-i ft Smell). 150 As we have already seen, his appearance 
here should probably be explained in the light of asdwira polemics; we may also 
have yet another example of the 'onomatomania' of the Islamic tradition. 1 l 





mia and Egypt (Albany: State University of New York Press, 1989), 227-9. For a rehabilitation 
m very different grounds) of the view that Jerusalem fell to the 'Israelites' because the latter 

penetrated the city's defences through an aqueduct, see Z. Abells and A. Arbit, 'Some new 
thoughts on Jerusalem's ancient water system', PEQ 127 (1995), 2. 

144 Noldeke, 'Chronik', 44 n. 1. • • , 

145 Tribesmen crowed about their presence at the battle, one boasting that he had participated 
in the battles of al-Qadisiya, Jalula', Tustar, Nihawand, and al-Yarmtik; see al-Fasawi (277/890), 
Kitcib al-ma'rifa wa-l-ta'nkh, ed. Akram Diya' al-'Umarl (Beirut: Mu'assasat al-nsala, 1981), 1, 
233. Cf. Ibn Qutayba (d. 276/889), 'UyBn al-akhhar, ed. Ahmad ZakI al-'Adawi (Cairo: Dar 

ib, 1343-48/1925-30), III, 245 (Isfahan, Tustar, Mihrajan, kuwar al-Ahwaz, Pars). 

Fa-qama Ziyad fa-takallama fa-ablagha fa-'ajiba al-itas mitt baydnihi wa-qalu inna Ibn 
' Ubayd la-khatlb; see al-Zubayri (d. 236/851), Nasab Quraysh, ed. E. Levi-Provencal (Cairo: Dar 
al-ma'arif, 1953), 244-5. The locus clmsims for Ziyad's eloquence is his famous khutba batra 
delivered to the Basrans; on his reputation for eloquence, see, Henri Lammens, "Ziad ibn AbThi, 
vice-roi de PIraq, lieutenant de Mo'iiwiya', reprinted in his Etudes sur k sieclc des Omayyadcs 
(Beirut: Imprimerie catholique, 1930), 60. 

147 Ibn Abi Shayba, Musannaf, VIII, 31-2; Khalifa ibn Khayyat, Ta rtkh, I, 138-42. Cf. also 
Ibn A'tham, Futilh, II, 18-25 (for heroes). 

148 Ibn Abi Shayba, Musannaf, VIII, 31. ,„..„. 
lAiJ ibid., VIII, 34: dihqaa Tustar, al-Tabarl, Ta'rTkh, I, 2554: rajul; Khalifa ibn Khayyat, 

Ta'nkli, I, 139: rajul min alii Tustar, al-Baladhun, Fundi, 380: rajulan inin al-a'ajim. 

150 Al-DInawarl, Akhbar, 138; al-Qummi, Tankh-i Qumm, 297-8. Cf. Ibn A'tham, Futuh, II, 
20, where 'NasTbeh' must be a variant of this name; also Ibn Abi Shayba, MusunnuJ, VIII, 28, 
where the traitor is identified as the brother of a victim of al-Hurmuzan. 

151 See Noth/Conrad, Early Arabic historical tradition, 126; also Crone, Slaves, 16. 

al-kutub, 1343 


,!'. 1 



Of the traitor's actual identity we shall probably never know the details, 152 
for the Nestorian authorities naturally had their own axes to grind; here, like 
in the Arabic, the identity of the traitor was polemically conditioned. The 
provenance of the Tustar traitor is suspiciously the same as that of a certain 
Peter, also a native of Bet Qatraye, who is said to have betrayed Alexandria 
to the Persians in an early part of the chronicle. 153 In neither Alexandria nor 
in Tustar can we corroborate the identities of these men, and to explain why 
Bet Qatraye is given to provide figures such as these we should probably look 
to the Nestorian ecclesiastical controversies that took place when our work 
was being assembled. For it was in the middle of the seventh century that 
the bishops of Fars, and soon after, Bet Qatraye, refused to acknowledge the 
authority of Isho'yab III, who served as catholicos of the Nestorian church 
from 649 to 659. 154 Several of the letters written by Isho'yab III address the 
problem of the recalcitrant bishops of Bet Qatraye, 155 and one, which can be 
dated to the period between 649 and 659, states that George, the bishop of 
Shush tra, was among those enrolled to argue the catholicos' view. 156 Just as 
in the case of the Islamic tradition, history was apparently pressed into service 
to express views about the present: the Qatarenes' threat to the unity of the 
Nestorian church in Isho'yab's day gave rise to the tradition of a Qatarene's 
betrayal of the Nestorians to the Muslims in Tustar. 

Our Syriac source cannot shed any direct light on a report that describes 
a sulk in Tustar, on which the Tustaris reneged (kafara); the city is then said 
to have been reconquered by muhajirun. 151 In its earliest datable form the 
tradition is credited by 'Abd al-Razzaq al-San'anl (d. 21 1/826), 158 as by 
al-Baladhun after him, to Ibn Jurayj (d. 150/767), on the authority of 'Ata' 
al-Khurasanl (d. 1 33/750). 159 The tradition being impossible to confirm, 160 we 
might explain it in the light of post-conquest polemics. Considering that the 
issue addressed by 'Ata' is a taxation anomaly — why 'Umar exempted the 
issue of conquest unions between the muhajirun and Tustarl women — one is 
tempted to think that the tradition is primarily aetiological. Similarly* if the 
purported participation of the muhajirun might have functioned to endow 
Tustar with high-status settlers, 161 so too might accounts that posit a city's 

152 There is no mention of a traitor in the account available to Ibn Sa'd {Tabaqat, V, 64), but 
here Ibn Sa'd is interested only in the events that follow al-Hurmuzan's surrender. 

153 See 25/22. On Bet Qatraye, see Jean Maurice Fiey, 'Dioceses syriens orientaux du Golfe 
Persique', in Memorial Mgi Gabriel Khouri-Sarkis (Louvain: Imprimerie Orientaliste, 1969), 
209-12 (reprinted in Communautes, Chapter II). 

154 For an overview of the controversy, see Fiey, 'Iso'yaw le Grand'. 

155 See Isovahb Patriarchae 111 Liber epistularum, ed. and trans. Rubens Duval (Paris: 
L. Durbecq, 1_904-1905; CSCO 11-12, Scr. syri 11-12), nos. 17-20 in the third cycle of letters, 
written while Isho'yab was catholicos. 

156 Liber epistularum, 259/187. 

157 Al-Baladhuri, Futuh, 382; Ibn Zanjawayh, Amwal, II, 439. 

158 'Abd al-Razzaq al-'San'anf, Musannaf, ed. HabTb al-Rahman al-A'zamT (Beirut: Al-Majlis, 
al-'ilml, 1 390- 1 407/ 1 970-87), V, 293 (first cited by Patricia Crone, 'The ifirst-cehtury concept of 
Higra\ Arabica 41 [1994], 358). 

159 On Ibn Jurayj and this 'Ata' (who is not to be confused with 'Ata' ibn Abl Rabah), see 
Harald Motzki, Die Anfdnge cier islamischen Jurisprudenz (Stuttgart: Franz Steiner, 1991), 183-218. 

160 To expect our Syriac source to concede that Tustar's Nestorian authorities reneged on an 
earlier agreement — unless, of course, it was to be portrayed as heroic resistance-^-is perhaps as 
unreasonable as it is to expect the Islamic tradition to record the apparently wanton killing of 
local Christians (on which see below). Hill {Termination, 134) is sceptical of this kufr tradition, 
suggesting that it refers to another (unnamed) city. 

' 61 See Noth/Conrad, Early Arabic historical tradition, 98, 210; and cf. Tarif Khalidi, Arabic 
historical thought in the classical period (Cambridge: Cambridge University Press, 1994), 46 
(explaining the chronological and geographical organization of the tabaqat): 'What may have 
been at issue is a kind of apostolic truth theory whereby the Prophet's companions and their 
descendants act as guarantors of the true faith in the cities where they settled'. (It almost goes 
without saying that the authors disagree about the reliability of the early source material.) Cf. 
C. F. Robinson, Islamic historiography (Cambridge: Cambridge University Press, 2003), 138 fT. 



kufr reconcile conflicting accounts of its conquest history. For while the con- 
quest tradition concedes the sulh ba'dfath arrangement, 162 it was too awkward 
to argue for afath bad sulh, since this would cast dishonour on the conquering 
Muslims: hence kufr accounts, which shift responsibility for renewing hostilities 
back to the conquered. 

If there is a kernel of truth in all of this, it is probably that the conquest 
was violent. That the Islamic tradition says nothing of the killing of local 
Christians is to be explained not only by its relative indifference to (and absence 
of solid information about) the fate of the conquered, 163 but also by the political 
circumstances in which it stabilized. Clearly defined legal rights and peaceful 
co-existence, the latter commonly articulated in the Prophetic prohibition of 
killing monks, 164 are developments of the post-conquest period. Of course a 
similar thing can once again be said about the Christian tradition: had our 
Syriac source been written a century later, when the Christian elites had begun 
to work out a modus vivendi with the Muslims, the killing might have been 
conveniently forgotten as well. 

Finally, an account that posits the discovery of an uncorrupted corpse of 
another (now unidentified) prophet in Tustar is almost certainly bogus. 163 It 
was probably invoked to support claims made in the course of the 'asabiyat 
that flared up between the Tustaris and Susls about Daniel's tabilt. 166 As a 
source of local pride, as well as a draw for pilgrims, sites such as these were 
obviously of some value. 167 

Tustar II: the organization of traditions 

For the purposes of historical reconstruction, we can say with some confidence 
that reports of a siege led by Abu Miisa al-Ash'arl, which was then followed 
by a betrayal from within, reflect early and authentic memories of the events 
in question. How was this memory transmitted? The question is a notoriously 
difficult one, but in Tustar we have enough evidence to tease out some 
provisional answers. 

We can start with the collections in which the Tustar accounts were 
included. The conquest traditions of Khuzistan seem to have been compiled 
into province-based collections (e.g. al-Mada'inl's 168 and Abu 'Ubayda's 169 
Futuh cd-Ahwaz), as well as into Basran-based collections (e.g. al-Mada'inl's 
Khabar al-Basra wa-futuhiha)} 10 Detailed descriptions of the first of these seem 
to be lacking in the literature, but we are fortunate to have a glimpse at the 
contents of the second. According to Ibn al-Nadlm, it began as follows: 
'Dastumaysan, the governorship of al-Mughlra ibn Shu'ba, the governorship 

162 Al-Tabar! Tankli, I, 2565; cf. al-Baladhun, Futuh, 378. 

163 For other examples of conquest killing, see Crone and Cook, Hagarism, 33. 

164 Thus Abu Yusuf, Kharaj, 195: ushab al-suwami'. 

165 Ibn Abl Shayba, Musanmf, VIII,' 31-2.' 

166 These are attested for a later period; see al-MuqaddasT (wr. c. 375/985), Afmm cd-taqaslm 
fima'rifat al-aqalTm, ed. M. J. de Goeje (Leiden: E. J. Brill, 1906; BGA 3), 417; also noted by 
Claude Cahen, 'Mouvements populaires et autonisme urbain dans l'Asie musulmane du Moyen 
Age', Arabica 6 (1959), 28. The Jews of al-Sus in Benjamin of Tudela's time are said to have 
argued about the tomb as well; see Benjamin of Tudela (fl. mid- 12th c), Itinerary, ed. and trans. 
Marcus Nathan Adler (New York: Philipp Feldheim, 1907), 52-3. 

167 In the thirteenth century Tustar could claim the tomb of the sixth Imam of the ShT'a, 
Ja'far al-Sadiq (d. 148/765); see al-HarawT (d. 611/1215), Al-Isharu ila mdrifat al-ziyara, trans. 
Janine Sourdei-Thomine as Guide des lieux de pelerinage (Damascus: Institut Francais de Damas, 
1957), 222-3. 

l " Ibn al-Nadlm, Fifuist, 103; Yaqut, hshdd, V, 316. 

169 See above, n. 58. 

170 Ibn al-Nadlm, Fihrist, 103. 


! J 



of Abu Musa, the matter (kfiabar) of al-Ahwaz, of al-Manadhir, of Nahr TTra, 
of al-SOs, of Tustar, 171 of the citadel {al-qal'a), of al-Hurmuzan, of Dabba ibn 
Mihsan, 172 of Jundaysabiir'. Why the material was assembled into this form, 
in addition to the more conventional Futuh al-Ahwaz form, can be explained 
at least in part by the administrative controversy that pitted Basrans against 
Kufans; for what we really have is a set of traditions recounting the victorious 
march of Basran armies against the remnants of the Sasanian state. 

There are, in addition, two very striking features in Ibn al-Nadlm's survey 
of al-Mada'inl's work. The first is that the order of titles — here representing 
'section headings' — clearly reflects the sequence of events and battles as they 
are known to us from (most of) the surviving sources: Abu Musa follows 
al-Mughlra ibn Shu'ba, and his appointment is followed by the conquests 
of al-Manadhir, Nahr TTra, al-Sus, and Tustar (Jundaysabur being misplaced 
after Tustar). 173 Given the dearth of second- and early third-century material, 
it is useful to know that the hard work of establishing a more or less correct 
sequence was apparently finished by this time. 174 

The second striking feature is the detail concerning the conquest of Tustar, 
particularly al-Hurmuzan's role in it. 175 Now in his attention to al-Hurmuzan, 
al-Mada'inl is clearly reflecting broader trends: thus Ibn AbT Shayba has a 
long section on 'What was related concerning Tustar' {ma dhukira fT Tustar), 
which is dominated by al-Hurmuzan, and the otherwise laconic Khalifa ibn 
Khayyat, drawing on sources that include al-Mada'inl, pauses for four pages 
of material on waq'at Tustar, here too al-Hurmuzan plays the starring role. 176 
What makes Ibn al-Nadlm's description of al-Mada'inl's, work especially 
interesting is his organization of this material into three discrete sections, i.e. 
khabar Tustar, khabar al-qal'a, and khabar al-Hurmuzan. The khabar al-qal'a 
must refer to a set of traditions concerning the siege of the city in general and 
al-Hurmuzan's sheltering inside the citadel (qal'a, qasaba) in particular; this is 
usually, but not always, described as the result of the Muslims' penetration of 
the city walls. The khabar al-Hurmuzan, it follows, would have been a collection 
of reports relating his surrender and meeting with 'Umar in Medina; a favourite 
account is a ruse by which al-Hurmuzan secured safe passage. 177 The concerns 
here are fairly easy to discern: to contrast the pious austerity of 'Umar with the 
imperious ostentatiousness of al-Hurmuzan — that is, to give vivid illustration 
to the Arabian God's victory over the polytheist Sasanians. 178 The dominant 
metaphor seems to be al-Hurmuzan's fine clothing, which is contrasted 
with 'Umar's spare garb; that the scene is a topos is almost certain. 179 This 

171 Fliigel {Fihrist, 103) here read Dastawa, which makes enough sense (see Yaqut, Mu'jam, II, 
574); but I follow Dodge {The Fihrist, I, 225) and Tajaddud {Fihrist, 115). 

172 See Ibn A'tham, Futuh, II, 28-30; al-Tabari, Ta'rikh, I, 2710-13. 

173 The early and indecisive campaigns that go almost entirely unnoticed by our Syriac source 
were presumably embedded in the section on the governorship of al-Mughlra ibn Shu'ba. 

174 Khalifa ibn Khayyat, who had access to al-Mada'inl's work on al-Ahwaz {Ta'rikh, I, 140: 
qala Aim l-Hasan), may have had the good judgement to ignore his sequence when it came to 

175 Caetani {Annali dell' Islam, III, 908-09) may have been the first to note the crucial 
role played by al-Hurmuzan in the conquest accounts. The advice given by al-Hurmuzan to 
'Umar about the conquest of Isfahan is discussed by Albrecht Noth, 'Isfahari-Nihawand. Eine 
quellenkritische Studie zur fruhislamische Historiographie', ZDMG 118 (1968), 283-4. 

176 Note as well that Sayfs account as preserved by al-Tabarl revealingly begins with 
biographical material on al-Hurmuzan; see al-Tabarl, Ta'rikh, I, 2534. • 

177 Khalifa ibn Khayyat, Ta'rikh, I, 142; al-Baladhuri, Futuh, 381. 

178 Thus al-Tabarl, Ta'rikh, I, 2557-8; Ibn Sa'd, Tahaqat, V, 64-5: al-hamd li'lldh alladhi 
adhalla hadha wa-shTatahu bi-l-Fslam, etc. 

179 See, for example, al-Ya'qubl, Historiae, II, 163. In her article 'al-Hurmuzan' in EI 2 , III 
(Leiden: E. J. Brill, 1971), 586b, Laura Veccia Vaglieri concedes that al-Hurmuzan's 'arrival in 
Medina is described with a number of details that seem to bear a romantic stamp'. 



leaves us with the problematic reading of khabar Tustar, if it is correct, it 
probably refers to the campaigning that led up to the siege. 

Of course, that al-Mada'inl organized a mass of Tustar! traditions in this 
fashion in no way means that they were always so carefully distinguished. This 
is made plain by a contemporary, Ibn Abl Shayba, a muhaddith who does us 
the favour of citing relatively full isnads, and who also eschews the akhbarTs' 
practice of breaking up and rearranging akhbar. His first account of the 
battle of Tustar and its aftermath was transmitted from Qurad Abu Nuh 
('Abd al-Rahman ibn Ghazwan, d. 207/822), 180 and is ultimately credited to 
'Abd al-Rahman ibn Abl Bakra (d. c. 100/718). 181 The account seems to reflect 
a fairly naive stage of tradition building. It takes the reader through the siege, 
surrender, and al-Hurmuzan's meeting with 'Umar; and for all that it presents 
an edifying story, organized primarily around the dialogue, it is disarmingly 
vague: we have but a handful of characters, and no attempt to locate the 
events chronologically. It may reasonably be taken to represent one late first- 
or early second-century Basran tale of the conquest. Khalifa ibn Khayyat had 
access to the same account, which he too credits to Qurad Abu Nuh, now via 
an intermediary, 'AIT ibn 'Abd Allah. 182 Whereas Ibn Abl Shayba probably 
preserved this account in externa, Khalifa ibn Khayyat, here wearing an 
akhbarTs hat, gives us a highly abbreviated version. It too enjoys pride of 
place in Khalifa's presentation, but now the account is stripped of all but its 
essentials, and breaks off- when al-Hormuzan takes refuge in his citadel. The 
tradition has apparently begun to fragment, in this case according to the 
categories reflected in al-Mada'inl's work. 183 




One can only agree with Conrad that 'work that securely vindicates, rather than 
repudiates,' the historicity of early Arabic accounts is extremely difficult'. 184 
As I have tried to show, our Syriac passage can be handled in such a way 
so as to vindicate and repudiate. Since much of the preceding has also been 
fairly rough going, I shall conclude by restating more concisely, and briefly 
elaborating upon, my principal conclusions. 

1, A local seventh-century Syriac source, which is historiographically independent 
of the Islamic tradition, can offer impressive corroboration for accounts pre- 
served in a range of Arabic-Islamic sources, which generally date from the 
ninth and tenth centuries. Since the corroboration is occasionally detailed and 
precise, in this case there can be no doubt that the nascent historical tradition 

180 Ibn Sa'd, Tabaqat, VII.2, 77; Ibn Hajar (d. 852/1449), Tahdhib a!-taluHub (Hyderabad: 
Da'irat al-ma'arif al-nizamlya, ah 1325-27), VI, 247-9; al-DhahabT (d. 748/1348), Siyar a'lam 
al-mibala, ed. Shu'ayb al-Arna'ut et al. (Beirut: Mu'assasat al-risala, 1401-04/1981-84), IX, 
518-19; al-Safadl (d. 764/1362), Al-WafT bi-l-wafaydt, ed. Helmut Ritter, Sven Dedering et al. 
(Istanbul and Wiesbaden: Deutsche Morgenlandische Gesellschafl, 1931-proceeding), XVIII, 217. 
,V W See al-Baladhuri, Futufi, 347; Ibn Sa'd, Tabaqat, VII. 1, 138; Ibn Qutayba, Kitdb al-ma'arif, 
ed. Tharwat 'Ukkasha (Cairo: Wizarat al-thaqafa, 1960), 289; Ibn Khallikan, Wafaydt, VI, 366; 
Ibn Hajar, Tahdhib, VI, 148-9; al-Safadl, Wajt, XVIII, 128; al-DhahabT, Siyar a'lam al-mtbald', 
IV, 319-20; Ibn al-'Imad (d. 1089/1679), Shadkarat al-dlutliab, ed. 'Abd al-Qadir al-Husnl 
al-Jaza'irl (Cairo: Maktabat al-qudsT, ah 1350), I, 122. Al-Wiiqidi also drew on 'Abd al-Rahman 
for information about al-Basra; see al-Tabarl, Ta'rikh, I, 2530. 
i . l82 Khalifa ibn Khayyat,* Ta'rikh, I, ' 1 39-40. 

\y.<. l83 Note the narrative interruption (gala) that may mark the division between khabar al-qal'a 
and khabar al-Hurmuzan material in Ibn AbT Shayba, Musannaf, VIII, 29:-4. 

184 Conrad, 'Arvvad', 399 n. 213. 



was in some measure continuous. The results here thus contrast sharply with 
another recent comparison of Arabic and Syriac sources, where it was shown 
that the former retain only the vaguest outlines of the conquest of Arwad, a 
small island off the coast of Syria. Here radical discontinuity was the lesson 
learned. 185 

Part of the explanation for the contrast may lie in the relative strengths of 
the Syrian and Iraqi historical traditions. For although Syria did produce more 
historiography than has generally been assumed, it cannot compare with that 
of Iraq; and what was produced in Syria was frequently slighted by later Iraqi 
authorities in favour of Iraqi traditions. 186 But since the invention of tradition 
was apparently not limited to Syria, 187 and furthermore, since the survival of 
some authentic material from Syria was occasionally possible as well, 188 this 
explanation cannot take us terribly far. It is thus probably more fruitful to 
draw a slightly different contrast. Left in the hands of the Iraqis, for whom 
the fate of the Mediterranean island of Arwad could hardly have constituted 
a serious concern, such conquest tradition as there was disintegrated almost 
entirely. 189 By contrast, we have seen that the conquest of Khuzistan in general, 
and Tustar in particular, mattered a great deal to the neighbouring Basrans 
and Kufans; 190 indeed, were it not for the Kufan/Basran debates, much more 
material might have been lost. It may seem trite to point out that history that 
matters is more readily transmitted than history that does not; but in this case 
it bears repeating. If we assume that the tradition remained oral beyond the 
lifetime of the participants, as we must, 191 the continuing interests of the 
Basrans and Kufans in the conquest fate of cities to the south provide the best 
explanation for the survival of material in oral form. There is no general life 
expectancy for oral traditions. 192 

185 ibid., particularly 388: '...the fact remains that it can be demonstrated in every case that 
the Arab-Islamic material for the conquest of Arwad does not and cannot consist of accounts 
passed on from one generation to the next in a continuous tradition beginning with the generation 
of the Arab conquerors. Instead, the beginnings of the extant tradition for this event must be 
sought among Umayyad storytellers piecing together narratives with only the barest shreds of 
genuinely historical information to guide or restrain the process of reconstruction'. 

186 See Fred M. Donner, 'The problem of early Arabic historiography in Syria', in Muhammad 
'Adrian al-BakhTt, ed., Proceedings of the second symposium on the history of Bilad al-Shani during 
the early Islamic period up to 40 A.H./640 A.D. (Amman: University of Jordan, 1987), I, 1-27. 
On a Damascene tradition, see Gerhard Conrad, Abul-Husain al-Razi (-347(958) und seine 
Schriften. Untersuchungen zur friihen Damaszener Geschich'tsschreibung (Stuttgart: Franz Steiner 
Verlag, 1991). See now J. Lindsay, ed., Ibn 'Asakir and early Islamic history (Princeton: Darwin 
Press?, 2001. 

I8 'See Noth, 'Isfahan-Nihawand'; Donner, Conquests, 198-9 (on Buwayb); and now 
Noth/Conrad, Early Arabic historical tradition. 

188 See Donner, Conquests, 144 (al-Waqidl apparently corroborated by the Syriac tradition; 
there is no evidence that the latter depended on the Islamic). 

189 Note that it is the Syriac tradition, in the person of Theophilus of Edessa (d. 785), that 
quite naturally transmits a more believable version of events. 

190 Cf. Lecker's comments a propos of Abu 'Ubayda ('Biographical notes', 17): '...the 
conquests of the Sawad and f the ' neighboring Ahwaz were a kind of local history for the 
Basran A.'U.'. •' , t 

191 The case that the early tradition was written down earlier is occasionally asserted (see, 
most recently, Khalidi, Arabic historical thought, 14, 26-7), but it has not been demonstrated. 
Much as one would like to see early Islamic scripturalism function as a catalyst for historical 
writing (cf. the role of Christianity in the shift from roll to codex), we lack the evidence to see 
this at work. For two recent views on the problem of the origins of Islamic historiography, see 
F. M. Donner, Narratives of Islamic origins: the beginnings of Islamic historical writing (Princeton: 
Darwin Press, 1 998) and Robinson, Islainic historiography. 

192 On the social function of oral history, see John Kenyon Davies, 'The reliability of oral 
tradition', in L. Foxhall and J. K. Davies, eds, The Trojan War: its historicity and context 
(Bristol: Bristol Classical Press, 1984), 90; O. Murray, 'Herodotus and oral history', in Heleen 
Sancisi-Weerdenburg and Amelie Kuhrt, eds, Achaemenid history, II: the Greek sources (Leiden: 
Nederlands Instituut voor het Nabije Oosten, 1987), 99. 



Of course, if the transmission of historical memory was not completely 
discontinuous, it was anything but disinterested. It is abundantly clear that 
much of the material was conditioned, and in some cases generated, by post- 
conquest polemics about spoils and administration. Even a tradition that at 
first glance suggests only simple storytelling of the awoil variety, e.g. 'the first 
to light the fire at the gate of Tustar', 193 is adduced by Abu Yusuf in his 
discussion of the division of spoils. 194 One can disagree with Dennett's 
qualifications of Becker, or Noth's qualifications of Dennett, but there is no 
denying the insight that draws together all their work, and which Calder has 
emphasized: conquest accounts 'should be recognized as bearers of ideological 
and juristic messages'. 195 To take only one example: if one's share of the booty 
was determined in part by whether one was walking rather than riding, 
and if the latter, on what kind of mount, 196 how are we to describe how 
such-and-such a city was taken? 

What our Syriac source shows, however — and this needs to be 
emphasized — is that the Khuzistan tradition is more than the accumulation of 
details arbitrarily added by storytellers, more than topoi and schemata, and 
^'finally more than back-projected legal precedents or assertions of state and 
provincial power. All of these do appear, crowding, and no doubt occasionally 
crowding out, authentic material. But some authentic material did survive, and 
since some of this at first appears to be manifestly stereotypical, the task of 
distinguishing between authentic and unauthentic is no simple matter. The 
conquest of Tustar shows many of the signs that usually betray literary effect, 
e.g. statements describing the enemy's strength, 197 a great siege, tribal boasting, 
and eschatological allusions, but for all these it cannot be dismissed as merely 

2. The survival of authentic material is most striking at the level of individual 
scenes (e.g. the siege/betrayal at Tustar; Daniel's tomb at al-Sus), although it 
is certainly true that legendary elements can arise here too (e.g. the traitor's 
name at Tustar). The results thus support Noth's view that the conquest 
traditions as we have them are generally composite reconstructions, assembled 
out of discrete units, rather than pieces of a now-lost coherent whole. 198 

This said, our Syriac source can also corroborate the Islamic tradition on 
matters that are not 'scene-specific', but rather represent a more synthetic 
understanding of events, e.g. the principal role played by Abu Musa al-Ash'an 
in the protracted campaigns, and matters of sequence as well, particularly the 
secondary capitulations of al-Sus and Tustar. This, in turn, suggests that at 
least some accounts concerning Tustar and al-Sus were integrated early on 
into a fairly broad view of the Khuzistan campaign, that the collectors and 
systematizers of the second and third centuries had the historiographical 
resources and sophistication to overcome the limitations of source material 
that did not, or some combination of both. It is the nature of our evidence — 
and the state of research — that we cannot say much more than this. One can 

193 See above, n. 148. 

194 Aba Yusuf, Kharaj, 198. 

195 Norman Calder, Studies in early Muslim jurisprudence (Oxford: Clarendon Press, 1993), 151. 

196 Al-Tabarl, Ta'rikh, I, 2556; Abu Yusuf, Kharaj, 18; 'Abd al-Razzaq, Musannaf, V, 183-7; 
Qudama ibn Ja'far (d. c. 310/922), Kitdb al-khardj, ed. Husayn Mu'nis (Cairo: Dar al-shuruq, 
1987), 59. 

™ n See, for example, al-Baladhun, Futuh, 380: wa-biha shawkat al-'aduw wa-hadduhum. 

198 Noth/Conrad, Early Arabic historical tradition, 5. On akhbar more generally, see Stefan 
Leder, 'The literary use of the khabar: a basic form of historical writing', in Cameron and 
Conrad, eds, The Byzantine and early Islamic Near East, 1, 277-315. 



speculate that the memory of Abu Musa was kept alive by descendants in 
al-Basra and al-Kufa; 199 and we have also seen that the correct sequence of 
battles was already in place by the early third century. But these are just two 
pieces of a much larger puzzle. 

3. The case of Khuzistan offers yet another illustration of how the 'schools 
theory' of the early tradition fails us. 200 If Ibn A'tham more frequently seems 
to have got things right, no single authority either resisted the forces of 
distortion completely, or monopolized early material entirely. In some cases 
the consensus of the Islamic tradition was vindicated; in others (e.g. the 
conquest of Jundaysabur), minority views were corroborated. Sayf ibn 'Umar 
seems to have been mistaken about the role of Abu Sabra at Tustar; 201 on the 
other hand, he seems to have been the only authority who had reasonably 
good material on the truce(s) between al-Hurmuzan and the campaigning 
Muslims. Indeed Sayfs account, which describes the tribute arrangements 
between al-Hurmuzan and the Muslims in an impressively imprecise way, 
passes Noth's standards for authenticity with flying colours. 202 The absence 
of detailed tribute accounts is an altogether striking characteristic of the 
Khuzistanl conquest accounts in general, and this too seems to be the case for 
all of our traditionists, regardless of their provenance. 

4. As far as the reconstruction of conquest history is concerned, we can have 
some confidence that Abu Musa al-Ash'arl, then based in al-Basra, led a 
Muslim force that followed up earlier battles in Khuzistan: the capitulations 
of ai-Sfis and' Tusfar. wtiicn we can actually describe in some detail", marked 
the turning point in his campaign. That the Sasanian defence and Muslim 
advance concentrated on these cities can be explained by their administrative 
significance in the late Sasanian period. 203 At least one truce was brokered, 
and as others preserved in very early sources, 204 it was apparently negotiated 
by commanders on the scene; it stipulated the payment of tribute and described 
a frontier. Our source cannot corroborate the Islamic tradition in dating 
matters, but it gives no reason to doubt that Tustar had fallen by 22 or 23 AH. 

199 See Ibn Hazm (d. 456/1064), Jamharat ansab al-'arab, ed. 'Abd al-Salam Muhammad 
Hariin (Cairo: Dar al-ma'arif, 1977), 397-8; Ibn Sa'd, Tabaqat, VI, 9. The significance of family 
and chin traditions is emphasized by Michael Lecker, The death of the Prophet Muhammad's 
father: did WaqidT invent some of the evidence?', ZDMG 145 (1995), 1 1. On family traditions in 
a different oral tradition, see Rosalind Thomas, Oral tradition and written record in Classical 
Athens (Cambridge: Cambridge University Press, 1989), chapter 2. 

200 The argument for distinct historiographical schools was undercut by Albrecht Noth long 
ago; see his 'Der Charakter der ersten grossen Sammlungen von Nachrichten zur fruhen 
Kalifenzeit', Der Islam 47 (1971), 168-99. 

201 Cf. the case of Abu 'Ubayda ibn al-Jarrah (Albrecht Noth, 'Fi/rw/;-history and Futuli- 
historiography', Al-Qantara 10 [1 989], 459), who seems to appear in Damascus conquest accounts 
onty to function within the manifestly late sulh/'anwa paradigm. 

- 02 'Je weniger eine Abgabe Steuercharakter hat, umso eher kann sie als authentisch angesehen 
wcrden; je mehr sie einer Steuer ahnelt, umso mehr ist ihre Authentizitat zu bezweifeln.' See 
Albrecht Noth, 'Die literarisch iiberlieferten Vertriige der Eroberungszeit als historische Quellen 
fur die Behandlung der unterworfenen Nicht-Muslime durch ihre' neuen muslimischen 
Oberherren', in Tilman Nagel et a/., eds, Studien sum Minderheitenproblem im Islam, I (Bonn: 
Selbstverlag des Orientalischen Seminars der Universitat Bonn, 1973), 300. For a balanced view 
of Sayf, see Ella Landau-Tasseron, 'Sayf ibn 'Umar in medieval and modern scholarship', Der 
Islam 67 (1990), 1-26. 

203 See Rika Gyselen, La geographic administrative de ('empire sassanide: les temoignages 
sigillographicptes (Paris: Groupe pour l'etude de la civilisation, du Moyen-Orient, 1989), passim; 
J. Markwart, A catalogue of the provincial capitals of Eranshahr (Rome: Pontificio Istituto Biblico, 
1931), 19. 


See ps.-Sebeos, Histoire d'Heraclius, 147, 164. 




Immediately following upon the passages I have translated above is an account 
alluding to the conquest of Syria and Egypt, and for the benefit of those 
interested, the following is a translation. 205 

Afterwards (batarken) a man from the Arabs named Kaled came and went 
to the West, and took the lands and towns as far as 'Arab. 206 Heraclius, 
the king of the Byzantines, heard [this] and sent a large army against them, 
whose leader was called S-q-y-1-r-a. 207 The Arabs defeated them, annihilating 
more thanl 00,000 Byzantines, whose commander they [also] killed. They 
also killed Isho'dad. the bishop of HIrta, who was there with 'AbdmasTh; 208 
this [Isho'dad] was undertaking an embassy between the Arabs and 
Byzantines. The Arabs [thus] took control of all the lands of Syria and 
Palestine. They wanted to enter the Egyptian [lands] as well, but they were 
unable, because the border (llwma) was guarded by the Patriarch of 
Alexandria with a strong and large army. For he had blocked the marches 
of the land, 209 and had built walls along the banks of the Nile in all the 
land. Only with difficulty, because of their (i.e. the walls') height, 210 were 
the Arabs able to enter and take the land of Egypt, Thebaid, and Africa. 

If only because of a possible allusion to the enigmatic al-Muqawqis, this 
passage deserves some attention. 211 


205 The passage begins on 37: 15/31:3 and ends on 38: 3/31: 20. 

206 Often glossed as western northern Mesopotamia under Byzantine rule; see Noldeke, 
'Chronik', 14 n. 4; Synodicon orientals, ou receuil de synodes nestoriens, ed. and trans. J.-B. Chabot 
(Paris: Academie des Inscriptions et Belles-Lettres, 1902; Notices et extraits des manuscrits de la 
Bibliotheque Nationale, 37), 617. 

207 37: 19, which is to be compared with Noldeke's and Brock's reconstruction of S[ac[ella]arius] 
in what is called the 'record dated to ad 637' in Palmer, Seventh century, 3; and Theophanes, 
Chronographia, ed. Karl de Boor (Leipzig: B. G. Teubner, 1883-85), am 6125: sakellarios; trans. 
C. Mango and R. Scott, The Chronicle of Theophanes' confessor (Oxford, 1997), 468f. For 
discussion see Donner, Conquests, 145-6; Walter Kaegi, Byzantium and the early Islamic conquests 
(Cambridge: Cambridge_University Press, 1992), 99-100. 

208 The presence of Isho'dad in Syria is curious, and it may be that this sentence is out of 
place; Fiey ('L'Elam... (suite),' 137), seems to put this episode of killing in Tustar. In 'AbdmasTh 
we almost certainly have 'Abd al-MasTh ibn 'Amr/'Amr ibn 'Abd al-MasTh, an AzdT native of 
al-HIra, who is well attested in the Islamic tradition: see al-Baladhun, Fulfill, 243; and Donner, 
Conauests, 183, 331 n. 83, for more literature. 

Literally: 'the entrances and exits'. 

210 Cf. the accounts beginning at 30: 25/26: 15. Walls were generally seen as an effective defence 
against Arabs (in contrast to seige-laying imperial armies); see Procopius (wr. 550), The history 
of the wars, ed. and trans. H. B. Dewing (Cambridge, MA: Harvard University Press, 1961 
reprint), II.xiv.12; and 'Joshua the Stylite', The Chronicle of Joshua the Stylite, 63/54. 

211 The testimony of the Khuzistdn chronicle is noted in the revised edition of Butler 
(Alfred J. Butler, The Arab conquest of Egypt, ed. P. M. Fraser, 2nd ed. [Oxford: Clarendon 
Press, 1978], ix), but it did not make it into the text proper. On al-Muqawqis, see K. Ohrnberg, 
art. 'al-Mukawkis' in EI 1 , VII (Leiden: E. J. Brill, 1993), 511a-513a; and Butler, Conquest, 
Appendix C; on the Great Wall of Egypt, see Butler, Conquest, 197-8; and on the walls and 
fortifications in general, Wladislaw Kubiak, Al-Fustat: its foundation and early urban development 
(Cairo: American University in Cairo Press, 1987), 50-57.